The rise of arbitration has been one of the most significant develop­ments in civil justice. Many scholars have criticized arbitration for, among other things, “privatizing” or “delegating” the state’s dispute-resolution powers and allowing private parties to abuse those powers with virtual impunity. An implicit assumption underlying this critique is that civil procedure, in contrast to arbitration, does not delegate sig­nificant state power to private parties.

This Article challenges that assumption and argues that we can address many of the concerns about arbitration by drawing on civil procedure’s solutions to its own delegation problem. From summonses to sub­poenas to settlements, civil procedure pervasively delegates state power during ordinary civil litigation. With these delegations comes the potential for abuse. But rather than limit private parties’ access to dele­gated power before any abuse has occurred, civil procedure generally polices its delegations for abuse after the fact. It does so in three main ways: by rescinding delegated power, as in the appointment of discovery masters; by withholding enforcement from an exercise of delegated power, as in civil Batson; and by punishing abuse of delegated power, as in Rule 11 sanctions. Civil procedure’s delegation-policing doctrines allow the state not only to protect private parties from harm but also to avoid becoming complicit in private exercises of delegated power that offend important public values.

Arbitration’s delegations of state power present many of the same problems as civil procedure’s, and scholars have rightly criticized the current arbitration regime for essentially writing a blank check to pri­vate parties. But whereas most scholars have focused on restricting access to arbitration’s delegations by deeming broad categories of arbi­tration clauses unenforceable, this Article suggests adapting civil procedure’s delegation-policing doctrines for arbitration. Even if courts continue to enforce arbitration clauses more often than arbitration’s critics would prefer, they should police arbitration’s delegations more closely than the law now permits.


    1. Defining Delegation
    2. Delegated Procedural Powers
      1. Prosecution
      2. Discovery
      3. Settlement
    3. Doubting Delegation
      1. Party Subordinance
      2. Managerial Judges
    1. Abusing Delegated Procedural Powers
    2. Civil Procedure’s Delegation-Policing Doctrines
      1. Rescinding Delegations
      2. Withholding Enforcement
      3. Punishing Abuse
    1. Complicity Avoidance in Civil Procedure
    2. Complicity Avoidance Beyond Civil Procedure
    3. Complicity Avoidance, Not Due Process
    4. Defending Complicity Avoidance—and Delegation
      1. The Political Theory of Complicity Avoidance
      2. Balancing Complicity Avoidance and Delegation
    1. Delegating Procedure Redux
      1. The Delegation Critique Defended
      2. Beyond Enforceability
    2. Policing Arbitration’s Delegations
      1. Rescinding Delegations
      2. Withholding Enforcement
      3. Punishing Abuse



Private parties are performing an increasing number of functions that were once discharged by the government. Indeed, privatization has become so pervasive that some public law scholars have dubbed it the “new religion.” 1 E.g., Martha Minow, Public and Private Partnerships: Accounting for the New Religion, 116 Harv. L. Rev. 1229, 1229 (2003) [hereinafter Minow, Public and Private Partnerships]. While scholars have criticized this development on many differ­ent grounds, 2 For a recent example, see generally Jon D. Michaels, Constitutional Coup: Privatization’s Threat to the American Republic (2017) (criticizing privatization for under­mining separation-of-powers principles). one particularly prominent criticism is that privati­zation involves a problematic “delegation” of state 3 For clarity, this Article uses “state” when referring to “the state” and “State” when referring to one of the fifty United States. power to private parties. 4 See generally Gillian E. Metzger, Privatization as Delegation, 103 Colum. L. Rev. 1367 (2003) (analyzing the constitutional problems posed by the privatization of govern­mental functions).

Dispute resolution has hardly been immune from the privatization trend. One of the most significant developments in civil justice during the last few decades has been the mass exodus of private disputes from public courts to private forms of dispute resolution, particularly arbitra­tion. 5 For an overview of the controversy surrounding consumer arbitration, as well as empirical findings confirming its significance, see generally David Horton & Andrea Cann Chandrasekher, After the Revolution: An Empirical Study of Consumer Arbitration, 104 Geo. L.J. 57 (2015). Many scholars have criticized arbitration in terms that echo some of the criticisms of privatization more generally. In particular, some have espoused various versions of what this Article calls the “delegation cri­tique.” The gravamen of the critique is that the current arbitration regime delegates state power to private parties and allows them to abuse that power with virtual impunity. Judith Resnik has thus criticized arbitra­tion for “delegating,” “outsourcing,” or “privatizing” the judicial func­tion. 6 See Judith Resnik, A2J/A2K: Access to Justice, Access to Knowledge, and Economic Inequalities in Open Courts and Arbitrations, 96 N.C. L. Rev. (forthcoming 2018) [hereinafter Resnik, Access] (manuscript at 174) (on file with the Columbia Law Review) (referring to mandatory arbitration as a “delegation” of judicial power); Judith Resnik, Diffusing Disputes: The Public in the Private of Arbitration, the Private in Courts, and the Erasure of Rights, 124 Yale L.J. 2804, 2811 (2015) [hereinafter Resnik, Diffusing Disputes] (arguing that the current arbitration regime constitutes a form of “outsourcing”); Judith Resnik, Lawyers’ Ethics Beyond the Vanishing Trial: Unrepresented Claimants, De Facto Aggregations, Arbitration Mandates, and Privatized Processes, 85 Fordham L. Rev. 1899, 1900 (2017) (describing arbitration as one facet of a broader “privatization and relocation of process”); Judith Resnik, The Privatization of Process: Requiem for and Celebration of the Federal Rules of Civil Procedure at 75, 162 U. Pa. L. Rev. 1793, 1809 (2014) [hereinafter Resnik, Privatization] (dubbing the enforcement of mandatory arbitration clauses a form of “privatization”); Judith Resnik, Procedure as Contract, 80 Notre Dame L. Rev. 593, 619–21 (2005) [hereinafter Resnik, Contract] (using the term “outsourcing” to describe the current arbitration regime); Judith Resnik, Uncovering, Disclosing, and Discovering How the Public Dimensions of Court-Based Processes Are at Risk, 81 Chi.-Kent L. Rev. 521, 549 (2006) [hereinafter Resnik, Uncovering] (same). Other scholars have used similar language. See, e.g., Peter L. Murray, Privatization of Civil Justice, 15 Willamette J. Int’l L. & Disp. Resol. 133, 142 (2007) (criticizing the current arbitration regime as “a kind of delegation of the judicial function to private dispute resolvers”); Seana Valentine Shiffrin, Remedial Clauses: The Overprivatization of Private Law, 67 Hastings L.J. 407, 441 (2016) [hereinafter Shiffrin, Remedial Clauses] (connecting concerns about arbitration to “the more general anxiety about the privatization of legal functions”). David Horton has argued that the current arbitration regime impermissibly delegates Congress’s lawmaking powers to private parties. 7 See David Horton, Arbitration as Delegation, 86 N.Y.U. L. Rev. 437, 469 (2011) [hereinafter Horton, Delegation]. Katherine Stone has made a similar argument. See Katherine V.W. Stone, Employment Arbitration Under the Federal Arbitration Act, in Employment Dispute Resolution and Worker Rights in the Changing Workplace 27, 51 (Adrienne E. Eaton & Jeffrey H. Keefe eds., 1999) (criticizing the Supreme Court’s “expansive interpretation of the [Federal Arbitration Act] as a form of delegation of judicial and legislative authority to self-regulating communities”); Katherine Van Wezel Stone, Rustic Justice: Community and Coercion Under the Federal Arbitration Act, 77 N.C. L. Rev. 931, 961 (1999) [hereinafter Stone, Rustic Justice] (noting “the extreme degree of delegation to arbitration that is reflected in the [Supreme Court’s] recent [arbi­tration] jurisprudence”). A host of other scholars have worried that arbitration permits private par­ties to use procedure to effectively override, or opt out of, substantive law. 8 See, e.g., Margaret Jane Radin, Boilerplate: The Fine Print, Vanishing Rights, and the Rule of Law 33–52 (2013) [hereinafter Radin, Fine Print] (arguing that boilerplate language in contracts perpetrates a form of “democratic degradation,” effectively erasing state-created legal rights); Cynthia Estlund, The Black Hole of Mandatory Arbitration, 96 N.C. L. Rev. (forthcoming 2018) (manuscript at 125) (on file with the Columbia Law Review) (arguing that a mandatory arbitration clause “virtually amounts to an ex ante exculpatory clause, and an ex ante waiver of substantive rights that the law declares non-waivable”); J. Maria Glover, Disappearing Claims and the Erosion of Substantive Law, 124 Yale L.J. 3052, 3075 (2015) (arguing that the current arbitration regime gives corporations “the power, through contract, effectively to negate substantive law”); David S. Schwartz, Enforcing Small Print to Protect Big Business: Employee and Consumer Rights Claims in an Age of Compelled Arbitration, 1997 Wis. L. Rev. 33, 110–22 [hereinafter Schwartz, Small Print] (citing empirical evidence that predispute arbitration clauses are meant to, and do in fact, function as “prospective waivers of substantive rights”). Some scholars have even contended that arbitration involves the exercise of such important state powers that it constitutes “state action” subject to the strictures of the Constitution. 9 See, e.g., Edward Brunet, Arbitration and Constitutional Rights, 71 N.C. L. Rev. 81, 109–13 (1992) [hereinafter Brunet, Arbitration and Constitutional Rights] (arguing that commercial arbitration involves state action because courts’ enforcement of arbitration agreements and awards implicates constitutional rights); Richard C. Reuben, Constitutional Gravity: A Unitary Theory of Alternative Dispute Resolution and Public Civil Justice, 47 UCLA L. Rev. 949, 989–1017 (2000) [hereinafter Reuben, Constitutional Gravity] (arguing that contractual forms of alternative dispute resolution can constitute state action); Richard C. Reuben, Public Justice: Toward a State Action Theory of Alternative Dispute Resolution, 85 Calif. L. Rev. 577, 609–33 (1997) (same); Jean R. Sternlight, Rethinking the Constitutionality of the Supreme Court’s Preference for Binding Arbitration: A Fresh Assessment of Jury Trial, Separation of Powers, and Due Process Concerns, 72 Tul. L. Rev. 1, 40–47 (1997) [hereinafter Sternlight, Rethinking] (same); Maureen A. Weston, Universes Colliding: The Constitutional Implications of Arbitral Class Actions, 47 Wm. & Mary L. Rev. 1711, 1742–67 (2006) (arguing that class arbitration constitutes state action); see also Martha Minow, Alternatives to the State Action Doctrine in the Era of Privatization, Mandatory Arbitration, and the Internet: Directing Law to Serve Human Needs, 52 Harv. C.R.-C.L. L. Rev. 145, 154–56 (2017) (expressing sympathy for the state action argument). But see Peter B. Rutledge, Arbitration and the Constitution 127–69 (2013) (contesting arguments that contractual arbitration constitutes state action); Sarah Rudolph Cole, Arbitration and State Action, 2005 BYU L. Rev. 1, 40–41, 48 (same). Though reflecting somewhat different concerns, each of these criticisms ultimately ques­tions whether the state should be coercively enforcing the products of private dispute resolution without any meaningful scrutiny. 10 See infra notes 311–314 and accompanying text. Scholars have leveled similar criticisms against other forms of “procedural private ordering,” see Jaime Dodge, The Limits of Procedural Private Ordering, 97 Va. L. Rev. 723, 728 (2011), particularly “contract procedure”—contractual provisions that purport to alter the procedural rules that generally govern the resolution of private disputes in public courts. The most prominent example is Kevin E. Davis and Helen Hershkoff’s article, which criticizes contract procedure as a form of (interchangeably) “privatization,” “outsourcing,” and “delegation,” and connects that concern to the worry that contract procedure forgoes various “public goods” associated with adjudication. Kevin E. Davis & Helen Hershkoff, Contracting for Procedure, 53 Wm. & Mary L. Rev. 507, 512–13 (2011). For another example, see David H. Taylor & Sara M. Cliffe, Civil Procedure by Contract: A Convoluted Confluence of Private Contract and Public Procedure in Need of Congressional Control, 35 U. Rich. L. Rev. 1085, 1087 (2002) (discussing predispute procedural agreements and courts’ role in enforcing them). These criticisms apply to arbitration a fortiori.

While largely accepting the main thrust of the delegation critique, this Article challenges one of its key assumptions, with implications both for our understanding of civil procedure and for our policies toward arbi­tration. Scholars have rightly worried that the current arbitration regime delegates significant coercive power to private parties and that the Supreme Court has construed the Federal Arbitration Act (FAA) so as to deny courts adequate authority to review arbitral proceedings and awards. 11 See infra section IV.A.1. For one example, see Hall St. Assocs. v. Mattel, Inc., 552 U.S. 576, 586–88 (2008) (holding that courts may not expand the grounds for modifying or vacating an arbitral award beyond those provided in the FAA, even with the agreement of the parties). The result is a system in which private parties and their chosen arbitrators enjoy wide latitude to abuse delegated state power, violating important public values in the process. Yet proponents of the delegation critique tend to assume that civil procedure, in contrast to arbitration, vests state power largely with state officials and institutions. 12 See infra notes 31–33 and accompanying text. In fact, this Article shows, civil procedure has already worked through its own delega­tion problem. Drawing on liberal political theory, this Article contends that broad delegations of state power are ubiquitous in ordinary civil lit­igation too. With these delegations comes the potential for abuse. Rather than limit private parties’ access to its delegations before any abuse has occurred, civil procedure primarily relies on a wide array of doctrines to police the delegations for abuse after the fact. 13 Much obviously turns on the meaning of the terms “police” and “abuse.” This Article elaborates on both at infra Part II. These delegation-policing doctrines enable the state to withhold or withdraw its support from exer­cises of delegated power that offend especially important pub­lic values. In revealing the extent of the delegations in ordinary civil liti­gation and examining civil procedure’s strategies for policing them, this Article develops a novel theoretical account of civil procedure. It then uses that account to show how we might adapt civil procedure’s delegation-polic­ing doctrines for arbitration. Civil procedure, this Article suggests, already contains the necessary conceptual and normative resources to address many of the problems with the current arbitration regime.

The claim that ordinary civil litigation pervasively delegates state power conflicts with commonly held impressions of our civil justice sys­tem. 14 See, e.g., Zachary D. Clopton, Redundant Public-Private Enforcement, 69 Vand. L. Rev. 285, 323 (2016) [hereinafter Clopton, Redundant] (“Courts actively supervise litigants and attorneys . . . .”); Scott Dodson, Party Subordinance in Federal Litigation, 83 Geo. Wash. L. Rev. 1, 20 (2014) (asserting that the circumstances in which litigants can act free from judicial oversight are “relatively rare and quite narrow”). In fact, civil litigation is a public institution administered largely by private parties. It is public, of course, in that it proceeds under the auspices of state bodies (that is, courts), subject to the ultimate decision­making authority of state officials (that is, judges). At the same time, however, our civil justice system delegates significant coercive power to private par­ties. Private parties decide whether to hale others into court; they deter­mine which documents to demand and whom to depose during dis­covery; and they often end up rendering binding dispositions of ongoing lawsuits by deciding whether to settle a case and on what terms. 15 See infra section I.B. Many times judges lack formal authority to preempt parties’ decisions on these matters. Even when judges do have such authority, their hands often remain tied, at least initially, as a practical matter. 16 See infra section I.B. In many other legal systems, state officials closely control the investigation and prosecution of civil cases. 17 On the greater involvement of state officials at all stages of civil litigation in civil law systems, see generally Mirjan R. Damaška, The Faces of Justice and State Authority: A Comparative Approach to the Legal Process (1986); Kuo-Chang Huang, Introducing Discovery into Civil Law 37–103 (2003). So too in our own criminal justice system, the investigatory and prosecutorial functions are largely the province of state police offi­cers and prosecutors. One of the hallmarks of our civil justice system, by contrast, is that it assigns these and other powers to private parties rather than state officials. 18 While other scholars have not completely ignored civil procedure’s various dele­gations, they have tended to view the delegations in isolation from one another rather than as manifestations of a single phenomenon. For example, Richard Nagareda depicted the power of class counsel to negotiate class action settlements as a delegation of lawmaking power without analyzing the problem of delegation in civil procedure more generally. See Richard A. Nagareda, The Preexistence Principle and the Structure of the Class Action, 103 Colum. L. Rev. 149, 191–98 (2003). More recently, Seth Davis has argued that, in certain cases, standing doctrine delegates to private parties the power to “enforce the law.” See Seth Davis, Standing Doctrine’s State Action Problem, 91 Notre Dame L. Rev. 585, 600–07 (2015). But in contrast to this Article, Davis denies that the delegation extends to ordinary private lawsuits and does not consider other state powers that civil procedure delegates to private parties. See id. Davis’s argument is discussed further below. See infra notes 68–75 and accompanying text. Along with comparativists and legal historians such as John Langbein, Stephen Yeazell has probably come the closest to offering a systematic account of the private exercise of power in civil litigation, though not the kind of theoretical framework developed in this Article. Yeazell’s work is discussed throughout Part I, infra.

As public law scholars have recognized, delegating state power to private parties creates a risk that the power will be abused. 19 See infra note 140 and accompanying text. The same risk exists in civil litigation, though the abuse assumes a different form in civil procedure than it does in the public law contexts on which scholars have primarily focused. While civil procedure generally abides a signifi­cant amount of abuse in the name of “litigant autonomy” and other com­peting values, it sometimes deems an exercise of delegated power too objection­able to tolerate. This Article shows that numerous procedural doctrines traditionally regarded as unrelated to one another actually per­form the same general function of policing civil procedure’s delegations for the most egregious forms of abuse.

Those doctrines perform this policing function using several differ­ent strategies, which can be arrayed along a spectrum according to how much they interfere with private parties’ exercise of delegated power. At one extreme, some doctrines rescind the delegation, putting state power back in the hands of state officials. Examples of this strategy include the appoint­ment of special masters to oversee discovery, the direct involve­ment of judges in settlement negotiations, and sua sponte dismissals. Other, subject-matter-specific examples include the right of the federal government to intervene in qui tam actions under the False Claims Act. 20 See 31 U.S.C. § 3730(b) (2012). At the other extreme, some doctrines punish only especially egregious abuses of delegated power after they have already occurred—either in the course of the litigation itself, as in the case of Rule 11 and discovery sanc­tions, or in collateral proceedings, as in the case of tort actions for malicious prosecution and abuse of process. In between are doctrines that leave the delegations intact but allow courts to withhold enforcement from specific exercises of delegated power that contravene public values. The foremost examples of this strategy are courts’ discretionary decisions refusing to compel discovery and quashing subpoenas, the constitutional ban on race-based peremptory challenges to prospective jurors in civil cases (so-called civil Batson), 21 See Edmonson v. Leesville Concrete Co., 500 U.S. 614, 631 (1991). and the line of procedural due process cases addressing State prejudgment attachment proceedings. 22 See infra note 198 and accompanying text.

This Article’s first contribution, then, is to theorize civil procedure’s response to the problem of delegated state power in ordinary civil litiga­tion. A second contribution is to explicate the normative logic underly­ing civil procedure’s delegation-policing doctrines. According to that logic, the delegation-policing doctrines perform a function that this Article calls “complicity avoidance.” The basic idea, immanent in the civil Batson case and other case law, is that when a private party abuses delegated state power, she thereby renders the state complicit in her wrongdoing; civil procedure’s delegation-policing doctrines allow the state to avoid this complicity by disavowing a particular exercise of dele­gated power that offends public values. The logic of complicity avoidance is rooted in liberal political theory, which undergirds the American constitutional order and its criteria for defining the legitimate exercise of political power. 23 This invocation of liberal political theory is orthogonal to the enduring debate over whether the U.S. Constitution’s ideological underpinnings are “liberal” or “republican.” See, e.g., Philip Hamburger, Liberality, 78 Tex. L. Rev. 1215, 1217–24 (2000) (reviewing the debate); Suzanna Sherry, Property Is the New Privacy: The Coming Constitutional Revolution, 128 Harv. L. Rev. 1452, 1458 n.21 (2015) (collecting sources on both sides of the debate). The answer is almost certainly “both.” See, e.g., Jack M. Balkin, Which Republican Constitution?, 32 Const. Comment. 31, 35 (2017) (reviewing Randy E. Barnett, Our Republican Constitution: Securing the Liberty and Sovereignty of We the People (2016)) (“The American constitutional tradition, understood in its best sense, has always drawn on elements of both the republican and liberal traditions . . . .”). Indeed, many scholars have argued that the American varieties of liberalism and republicanism are largely reconcilable. See, e.g., 1 Bruce Ackerman, We the People: Foundations 29–32 (1991) (defending “liberal republicanism”). In any event, the specific element of liberal political theory on which this Article focuses—its account of political power—is sufficiently ecumenical to accord with a wide range of constitutional theories. For further discussion, see infra section III.D.1. Most versions of liberalism hold that the state’s coercive power is collectively authorized by the members of the political commu­nity and exercised in their name. This creates the imperative for the state, acting on behalf of the political community, to repudiate objection­able exercises of delegated power by private parties during civil litigation. On this account, liberal political theory provides a strong justification for the current shape of our civil justice system.

The theory of complicity avoidance also connects civil procedure’s delegation-policing doctrines to other, substantive legal doctrines that are similarly concerned with ensuring that public power does not end up facili­tating objectionable private projects. The two foremost examples of the latter are Shelley v. Kraemer, which applied constitutional law’s state action doctrine to prohibit courts from enforcing racially restrictive covenants, 24 334 U.S. 1, 19–21 (1948). and contract law’s unconscionability doctrine, as reconcep­tualized by Seana Shiffrin. 25 See generally Seana Valentine Shiffrin, Paternalism, Unconscionability Doctrine, and Accommodation, 29 Phil. & Pub. Aff. 205 (2000) [hereinafter Shiffrin, Paternalism] (defending the unconscionability doctrine as a means by which courts can refuse to facilitate objectionable private conduct). This Article thus shows civil proce­dure’s delega­tion prob­lem to be but one manifestation of a much more pervasive problem that is addressed through substantive doctrines else­where in the law.

In exploring civil procedure’s delegations and its various strategies for policing them, this Article develops an analytical framework that bears on several pressing procedural problems. But here the focus is on the impli­cations for perhaps the most urgent one: arbitration. One of the Article’s main contentions is that the delegations in arbitration differ from the del­egations in ordinary civil litigation in degree rather than kind. 26 Other scholars have noted certain functional similarities between ordinary civil litigation and arbitration but not the specific similarity analyzed in this Article. Resnik has long criticized what she views as the encroachment of the contract-based norms of alternative dispute resolution (ADR) into ordinary civil litigation, through practices such as court-annexed ADR. See, e.g., Resnik, Diffusing Disputes, supra note 6, at 2806–07, 2844–50; Judith Resnik, Many Doors? Closing Doors? Alternative Dispute Resolution and Adjudication, 10 Ohio St. J. on Disp. Resol. 211, 212 (1995); Resnik, Privatization, supra note 6, at 1814; Resnik, Contract, supra note 6, at 622–23. This Article, by contrast, reverses the direction of the normative migration, conceiving of ADR as an extension of the delegations that have long been present in ordinary civil litigation.
In a similar vein, Richard Freer has recently argued that, with the demise of the civil trial, ordinary civil litigation has increasingly come to resemble arbitration, but in the specific sense that both now focus narrowly on the resolution of individual disputes and fail to realize the “broader goals and values” traditionally associated with adjudication, such as governmental transparency, democratic participation, and law generation. Richard D. Freer, Exodus from and Transformation of American Civil Litigation, 65 Emory L.J. 1491, 1492 (2016). Moreover, in contrast to this Article, Freer attributes this convergence to various historical contingencies rather than to any fundamental feature of civil procedure, such as its delegation of many state dispute-resolution functions to private parties. See id. at 1493.
And yet, many proponents of the delegation critique advocate policy proposals for arbitration that differ from civil procedure’s delega­tion-policing doctrines in kind rather than degree. They tend, in particular, to seek to restrict arbitration’s delegations, insisting either that courts deem broad categor­ies of arbitration clauses unenforceable, so as to channel many more disputes into public venues, 27 See, e.g., Radin, Fine Print, supra note 8, at 166–67, 173–74, 183; Horton, Delegation, supra note 7, at 441–43; Schwartz, Small Print, supra note 8, at 125–32; Jean R. Sternlight, Creeping Mandatory Arbitration: Is It Just?, 57 Stan. L. Rev. 1631, 1661–73 (2005) [hereinafter Sternlight, Mandatory Arbitration]; see also infra note 332 and accompanying text. or that arbitrators be required to employ court-like procedures, so as to conform to the proce­dural norms of public adjudication. 28 See, e.g., Resnik, Contract, supra note 6, at 600 (insisting that arbitral “processes have to look more like what courts do than not”); Reuben, Constitutional Gravity, supra note 9, at 1054–100; Sternlight, Rethinking, supra note 9, at 80–97; see also infra note 334 and accompanying text. These are reasonable responses, especially given the gross power asymmetries that produce the most controversial arbitration clauses. Nonetheless, even if courts leave arbitra­tion’s delegations largely intact, civil procedure’s delegation-policing doctrines suggest an alterna­tive approach that could render those dele­gations less troubling—more akin to the delegations we have come to accept in ordinary civil litigation.

Rather than restrict access to arbitration’s delegations before any abuse has occurred, such an approach would focus on policing exercises of delegated power for abuse after the fact. More specifically, a delega­tion-policing approach would address the abuse of delegated state power in arbitration by subjecting arbitrators’ decisions to more searching judi­cial scrutiny. That scrutiny could take a number of different forms, including allowing interlocutory review of key procedural decisions by arbitrators, more frequently granting motions to vacate (or more frequently refusing to confirm) arbitral awards that issue from proceed­ings marred by especially serious procedural or substantive defects, and even sanctioning parties who seek to confirm such awards. While some of these proposals would require amending the FAA, others could be implemented through more faithful judicial interpretations of the statute than the Supreme Court has rendered. To be sure, even robust ex post policing mechanisms would not solve all of the problems with arbitration; ex ante regulation might still be necessary to address some of the very worst abuses, such as class-arbitration bans. 29 See infra note 341 and accompanying text. But see Am. Express Co. v. Italian Colors Rest., 133 S. Ct. 2304, 2308–12 (2013) (holding that the FAA requires courts to enforce class-arbitration waivers); AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 351–52 (2011) (holding that the FAA preempted California’s ban on class-arbitration waivers). The point is that, even if arbitration clauses con­tinue to be enforced more often than arbitration’s critics think legitimate, both arbitral proceedings and awards should be subject to greater judicial scrutiny than current doctrine permits.

Although other scholars have also called for this kind of enhanced judicial review, they typically present it as only one among many indiscrim­inate limits on arbitration. 30 See infra notes 335–336 and accompanying text. What distinguishes a delegation-policing approach is that it sees ex post judicial review as a distinct strategy, one grounded in the same logic underlying the doctrines that respond to civil procedure’s own delegation problem.

The Article proceeds as follows. Part I makes the case for concep­tual­izing the exercise of various dispute-resolution functions by private parties in ordinary civil litigation as a delegation of state power. Part II then explains why civil procedure’s delegations need to be policed and examines the different procedural doctrines that fill that role. Those doctrines, Part III shows, perform a complicity-avoidance function analo­gous to that performed by various substantive doctrines—a function that cannot be understood in standard due process terms. Turning to the implications for debates about arbitration, Part IV identifies several ways in which the delegations in arbitration can be policed, even if not eliminated.

The Conclusion reflects on some of the Article’s methodological impli­cations. It contends that civil procedure scholars tend to underesti­mate the extent of civil procedure’s delegations and to exaggerate the differences between arbitration and ordinary civil litigation because of the discipline’s overreliance on the procedural due process framework, which is concerned with the procedures of state institutions rather than the exercise of state power by private parties. Greater attention to liberal political theory and its account of political power can help us to conceptualize many of the problems presented by the privatization of civil justice and to uncover solutions latent in civil procedure doctrine itself.

I. Civil Procedure’s Delegations

In criticizing arbitration for “privatizing” or “outsourcing” dispute resolution, proponents of the delegation critique assume a traditional conception of ordinary civil litigation as being part of “a rigidly indepen­dent public legal system.” 31 Dodson, supra note 14, at 45; see also Resnik, Diffusing Disputes, supra note 6, at 2806 (noting that typically, “[c]ourts are equated with public processes”). There are many different senses in which our civil justice system might be considered “public.” But for purposes of the delegation critique (and hence this Article), the most important is that “[c]ourts are seen as . . . proceeding under a system not of their users’ personal design but fashioned by bodies of rule-makers committed to procedural neutrality and subjected to public scrutiny.” 32 Resnik, Diffusing Disputes, supra note 6, at 2834. Civil litigation, on this view, is public insofar as it is administered by state institutions—courts—that, in turn, proceed according to state-made rules. 33 See, e.g., Owen M. Fiss, The Supreme Court, 1978 Term—Foreword: The Forms of Justice, 93 Harv. L. Rev. 1, 30–31 (1979) (arguing that a judge “is a public officer; paid for by public funds; chosen not by the parties but by the public or its representatives; and empowered by the political agencies to enforce and create society-wide norms”); Reuben, Constitutional Gravity, supra note 9, at 1048 (“Virtually all of th[e] activity [in ordinary civil litigation] takes place in the public courthouse and is presided over by a judge with plenary power to administer and decide the case as he or she deems appropriate.”); cf. David Marcus, From “Cases” to “Litigation” to “Contract”: A Comment on Stability in Civil Procedure, 56 St. Louis U. L.J. 1231, 1237 (2012) (identifying as the presumptively “legitimate” procedural form the “ideal-type” of the “individual lawsuit,” in which “the case proceeds before a detached, neutral judge in a courtroom open to the public”).

This Part aims to complicate that picture of ordinary civil litigation. The traditional conception actually conflates two distinct issues. The first is the nature of the functions performed in the course of resolving private disputes. 34 Cf. Metzger, supra note 4, at 1424 (arguing that an “inquiry into the nature of the powers exercised by private entities . . . is essential given the way that privatization not only enhances private actors’ power over others, but provides them with forms of authority conventionally understood to constitute government power”). On that issue this Article agrees with proponents of the delega­tion critique—and will argue in more detail in this Part—that binding “dispute-resolution services constitute an inherently govern­mental func­tion,” and indeed “an aspect of sovereignty.” 35 Davis & Hershkoff, supra note 10, at 535.

The second issue is who actually performs the various dispute-reso­lution functions involved in ordinary civil litigation. Given the scholarly rhetoric extolling the “public” nature of our civil justice system, one might be tempted to think that those functions are performed exclu­sively by state officials and institutions. 36 To be sure, civil procedure scholars have criticized certain aspects of contem­porary civil litigation, such as contract procedure and court-annexed ADR, as forms of “privatization.” But they depict these features as departures from, and corruptions of, ordinary civil litigation’s fundamentally public nature. See supra notes 10, 26. Scholars have, however, long recognized that civil procedure puts various powers in the hands of private parties rather than judges. 37 See, e.g., David A. Sklansky & Stephen C. Yeazell, Comparative Law Without Leaving Home: What Civil Procedure Can Teach Criminal Procedure, and Vice Versa, 94 Geo. L.J. 683, 684 (2006) (“Civil litigation . . . has been essentially privatized . . . .”). This Part argues that many of these private exercises of power in ordinary civil litigation should be concep­tualized as delegations of state power.

More specifically, this Part contends that the Federal Rules of Civil Procedure delegate state power to private parties in the following func­tional sense: At various points in the litigation, the Rules enable the par­ties to force the court to deploy coercive power for their own purposes. The Rules may technically interpose a state official in some of those instances, such that the parties cannot be said to formally wield coercive power, but the official nevertheless has little or no discretion to preempt the parties’ decisions, such that the parties effectively wield coercive power in the first instance. 38 This is different from the claim that the Federal Rules allow litigants to engage in “rulemaking,” in the sense of selecting the procedural rules that will govern their dispute. See Robert G. Bone, Party Rulemaking: Making Procedural Rules Through Party Choice, 90 Tex. L. Rev. 1329, 1342–52 (2012) [hereinafter Bone, Party Rulemaking]; see also Stephen C. Yeazell, Civil Procedure 138 (7th ed. 2008) (noting that many procedural rules are merely “default rules”); Robin J. Effron, Ousted: The New Dynamics of Privatized Procedure and Judicial Discretion, 98 B.U. L. Rev. 127, 141–44 (2018) (cataloging the various ways in which the Federal Rules defer to party “preference” or “agreement”). By contrast, this Article contends that the Rules sometimes allow litigants (either formally or effectively) to exercise the state’s coercive power, whether through “rulemaking” or otherwise.

Section I.A briefly elaborates this Article’s conception of a delega­tion of state power. Section I.B then examines three of the most notable delegations in civil procedure, while section I.C considers two general objections to conceptualizing any part of ordinary civil litigation as a del­egation of state power.

A. Defining Delegation

To determine when the state delegates power to private parties, we can use either a narrow, formal conception of a delegation or a broader, functional one. On the formal conception, the state delegates power when it authorizes a private party to act on its behalf with respect to other private parties, as when the government contracts with a private entity to administer a public benefits program. This is the prevailing con­ception among public law scholars 39 See, e.g., Metzger, supra note 4, at 1456, 1462–70 (arguing that the creation of an agency relationship is the most significant criterion for identifying a delegation of state power); Jed Rubenfeld, Privatization, State Action, and Title IX: Do Campus Sexual Assault Hearings Violate Due Process? 25 n.132 (Yale Law Sch., Pub. Law Research Paper No. 588, 2016), (on file with the Columbia Law Review) (consider­ing state action “cases in which government uses private parties for its own ends—i.e., when it delegates powers to private parties but continues to direct their objectives”). —and for understandable reasons. For one thing, “delegation” is a loaded (and, at least in public law con­texts, largely pejorative) term, so it makes sense to reserve that label for situations that are thought to raise the most serious concerns. For another, by cabining the category of delegations, we can avoid the sweep­ing implication that all power held by private parties is delegated state power, which would threaten to eliminate any distinction between state authority and private ordering. 40 See Metzger, supra note 4, at 1462 & n.326; see also infra note 47 (discussing the realist critique of the public–private distinction).

The functional conception of a delegation, by contrast, does not look for an agency relationship between the state and a private party but rather focuses on who controls the application of coercive power as a practical matter, regardless of who holds it as a formal matter. On this view, a private party is delegated state power when she can prompt a state official or institution to coerce another private party without the official’s or institution’s first having an opportunity to assess the grounds on which the coercion is being exercised. Although the coercive power may for­mally be vested with the state rather than the private party, the private party alone determines its initial application, and so effectively holds the power in the first instance.

While we should appreciate the risks associated with expanding the notion of a delegation, it is nevertheless useful to conceptualize such pri­vately initiated exercises of coercion as delegations of state power. Only by doing so can we make sense of the delegation critique of arbitration, which presupposes the broader conception of a delegation. As we shall see, the current arbitration regime does not formally authorize private parties’ chosen arbitrators to reach binding resolutions of private disputes; rather, it effectively renders arbitrators’ decisions binding by requiring courts to enforce them without any meaningful review. 41 See infra section IV.A.1. Arbi­tration thus delegates (or “privatizes”) state power only in the functional sense.

The functional delegations, moreover, raise the same normative con­cerns as the formal ones. State power is distinguished by its coercive­ness, 42 See infra section III.D.1; see also, e.g., David M. Lawrence, Private Exercise of Governmental Power, 61 Ind. L.J. 647, 647–48 (1986) (identifying “governmental” powers in terms of the exercise of coercion). But cf. Lillian BeVier & John Harrison, The State Action Principle and Its Critics, 96 Va. L. Rev. 1767, 1785 (2010) (distinguishing state action according to “the presence or absence of a decisive choice by a private person,” rather than the presence or absence of coercion). Without wading into longstanding philosophical debates about the nature of coercion, this Article accepts that state power is coercive in the sense that the state “typically backs its commands and prohibitions with the credible threat of brute force or other sanctions in the event of noncompliance.” Frederick Schauer, The Force of Law 5 (2015). but the state can enable private parties to exercise coercion other than by formally making them its agents. That is perhaps no more true than in civil procedure. What makes the various dispute-resolution powers involved in civil litigation “inherently governmental” or “aspect[s] of sovereignty” 43 Davis & Hershkoff, supra note 10, at 535. is the fact that they are binding—immediately backed by coercion. 44 Cf. Boddie v. Connecticut, 401 U.S. 371, 375 (1971) (recognizing “the State’s monopoly over techniques for binding conflict resolution”). As Part III explains more fully, liberal political theory has tra­ditionally conceived of the state as possessing a monopoly on the legiti­mate use of coercion. And indeed, the next section shows that the coer­cive dispute-resolution powers in civil litigation are the kinds of powers that have traditionally been exercised by the state and that continue to be exercised by the state in other contexts. Yet the next section also iden­tifies a number of significant points in civil litigation at which private par­ties initially determine the application of coercive power. These privately initiated exercises of coercion test the limits of the traditional liberal con­ception of the state just as much as formal conferrals of authority to act on the state’s behalf do, and thus equally merit the “delegation” label. 45 By contrast, even on this functional account, extralegal, custom-based regimes of dispute resolution do not involve a delegation of state power. See generally, e.g., Robert C. Ellickson, Order Without Law: How Neighbors Settle Disputes (1994) (using disputes between cattle ranchers in Shasta County, California, to demonstrate how neighbors often rely on social norms, rather than the formal legal system, to resolve disputes); Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. Legal Stud. 115 (1992) (explaining how the diamond industry has established its own dispute-resolution system independent of any state-run legal system).

Some might worry that the functional conception of a delegation proves too much, implying that all de facto power held by one private party over another is delegated state power. While that might follow if the state delegated power simply by enforcing, or giving effect to, private parties’ choices, the functional conception of a delegation requires some­thing more. When the state makes itself available to coercively enforce certain private choices, it generally retains broad discretion to set the terms on which it will provide that support; think, for instance, of the intricate legal regime governing the enforcement of private contracts. 46 Cf. Margaret Jane Radin, Boilerplate: A Threat to the Rule of Law?, in Private Law and the Rule of Law 288, 300 (Lisa M. Austin & Dennis Klimchuk eds., 2014) [hereinafter Radin, Threat] (arguing that the state’s coercive “[e]nforcement of exchanges by private ordering” is legitimated by a “background legal infrastructure” that leaves the state with significant discretion to scrutinize the voluntariness and substance of private agreements). The examples considered in this Article go much further: As the next section shows, the Federal Rules often effectively require courts to exer­cise coercive power at private parties’ behest, denying them any meaning­ful discretion to withhold that coercion, at least as an initial matter. One can recognize those instances as delegations without having to maintain that the state delegates power whenever it supports private ordering. 47 Thus, even as this Article seeks to broaden the notion of a delegation, it still main­tains a distinction between “public” and “private”—a distinction legal realists have famously dismissed as illusory. For prominent contemporary versions of that criticism, see Louis Michael Seidman & Mark V. Tushnet, Remnants of Belief: Contemporary Constitutional Issues 61–67 (1996) (arguing that, given the conceptual confusion surrounding the public–private distinction, “the state action analysis ought to be abandoned altogether”); Cass R. Sunstein, The Partial Constitution 351–53 (1993) [hereinafter Sunstein, Partial Constitution] (discussing the state’s role in setting the legal baselines against which private ordering occurs). Notwithstanding the force of the realist critique, the public–private distinction remains an enduring feature of legal practice and scholarship and indeed a “foundational premise of our constitutional order.” Metzger, supra note 4, at 1369–70; see also, e.g., Radin, Fine Print, supra note 8, at 36 (“[O]ur society’s underlying commitment to the idea of private ordering, which is embedded in our legal infrastructure of contract, unequivocally relies on the existence of a distinction between the realms of public and private action and ordering—a public/private distinction.”); J.M. Balkin, Populism and Progressivism as Constitutional Categories, 104 Yale L.J. 1935, 1968–69 (1995) (reviewing Cass R. Sunstein, Democracy and the Problem of Free Speech (1993)) (suggesting that, rather than “abolish the distinctions between concepts like public and private,” we should “understand these boundaries as more flexible”); cf. Jody Freeman, The Private Role in Public Governance, 75 N.Y.U. L. Rev. 543, 551 (2000) (conceding that “the particular coercive power of the state . . . is undeniable” even while endorsing the realist critique of the public–private distinction). That is perhaps no more apparent than in the delegation critique of arbitration itself, which presupposes fairly distinct public and private spheres and purports to distinguish between public and private exercises of state power. This Article thus assumes the distinction’s validity.

B. Delegated Procedural Powers

This section contends that three of the most significant aspects of ordinary civil litigation today—prosecution, discovery, and settlement—involve delegations of state power. At each of those stages, the Federal Rules effectively allow private parties to exercise coercive powers over others without prior judicial approval. There is a tendency, moreover, to deem those same powers “public” when exercised by state officials in other contexts; they remain public when exercised by private parties in the course of ordinary civil litigation.

1. Prosecution. — From the very outset of a lawsuit, the Federal Rules vest private parties with significant power. Rather than charge state offi­cials with prosecuting civil claims or require victims to vet their claims with state institutions before filing suit, the Rules generally authorize “aggrieved victims of wrongs to enlist the adjudicative machinery of the state in seeking redress.” 48 Sklansky & Yeazell, supra note 37, at 687. There are a number of notable exceptions to this general rule for certain kinds of claims. See David Freeman Engstrom, Agencies as Litigation Gatekeepers, 123 Yale L.J. 616, 644–55 (2013) [hereinafter Engstrom, Litigation Gatekeepers] (discussing instances in which administrative agencies serve as “gatekeepers,” screening legal claims before potential plaintiffs may proceed to court). Civil procedure scholars have, for the most part, failed to closely analyze the nature of the civil prosecutorial power, 49 An important, if limited, exception is the extensive body of scholarship on the “private enforcement” of public regulatory laws. See infra note 291 and accompanying text. but it has garnered more attention from proponents of the “civil recourse” theory of tort law. One of civil recourse theory’s main claims is that “[t]ort law . . . is about empowering private parties to initiate pro­ceed­ings designed to hold tortfeasors accountable.” 50 John C.P. Goldberg & Benjamin C. Zipursky, Torts as Wrongs, 88 Tex. L. Rev. 917, 946–47 (2010) [hereinafter Goldberg & Zipursky, Torts as Wrongs]; see also Benjamin C. Zipursky, Civil Recourse, Not Corrective Justice, 91 Geo. L.J. 695, 720–21, 735 (2003) [hereinafter Zipursky, Civil Recourse] (noting that defendants in tort suits are under no legal obligation to pay damages until a plaintiff has brought suit and secured a verdict in her favor). At times, Goldberg and Zipursky appear to argue that plaintiffs possess not just a prosecutorial power to initiate a lawsuit but also a remedial power to “exact” or take something from the wrongdoer should they prevail on their claims. See, e.g., Goldberg & Zipursky, Torts as Wrongs, supra, at 919; Zipursky, Civil Recourse, supra, at 718–19. The latter would arguably constitute another significant delegation of state power, but the claim is controversial, and this Article does not pursue it as an example of delegation in ordinary civil litigation. Replace “tortfeasors” with “wrongdoers,” and this claim actually describes a power enjoyed by all potential civil plaintiffs, not just tort plaintiffs. 51 See Arthur Ripstein, Private Wrongs 271–72 (2016) [hereinafter Ripstein, Private Wrongs]; John Gardner, Torts and Other Wrongs, 39 Fla. St. U. L. Rev. 43, 47 (2011); Scott Hershovitz, Harry Potter and the Trouble with Tort Theory, 63 Stan. L. Rev. 67, 102 n.105 (2010); Ori J. Herstein, How Tort Law Empowers, 65 U. Toronto L.J. 99, 109 (2015) [hereinafter Herstein, How Tort Law Empowers]; Gregory C. Keating, The Priority of Respect over Repair, 18 Legal Theory 293, 312 (2012). Indeed, civil recourse theorists have basically conceded the point, see, e.g., John C.P. Goldberg & Benjamin C. Zipursky, Civil Recourse Revisited, 39 Fla. St. U. L. Rev. 341, 348–54 (2011); Zipursky, Civil Recourse, supra note 50, at 734, and some scholars have attempted to extend civil recourse theory beyond tort law, see, e.g., Nathan B. Oman, Consent to Retaliation: A Civil Recourse Theory of Contractual Liability, 96 Iowa L. Rev. 529, 531–34 (2010) (developing a civil recourse account of contract law). More precisely, Ori Herstein, a sympathetic critic of civil recourse theory, has explained that a civil plaintiff’s ability to initiate a lawsuit includes both (1) a Hohfeldian power to subject another private party to the judicial process and to the courts’ authority to alter the parties’ legal rights and (2) a Hohfeldian privilege to decide whether to exercise that power. 52 Herstein, How Tort Law Empowers, supra note 51, at 104, 109; see also Zipursky, Civil Recourse, supra note 50, at 741 (“A right of action is a privilege and a power . . . .”). See generally Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning (Walter Wheeler Cook ed., 1919) (distinguishing between various juridical relations, including privileges and powers). That combination means that private parties, rather than state officials, generally get to decide who will be subject to the courts’ authority to adjudicate civil claims.

The power to initiate a lawsuit, while significant, is not in itself coer­cive. But the Rules go a step further, backing that power with an addi­tional, coercive power. Rule 4 provides: “On or after filing the complaint, the plaintiff may present a summons to the clerk for signature and seal. If the summons is properly completed, the clerk must sign, seal, and issue it to the plaintiff for service on the defendant.” 53 Fed. R. Civ. P. 4(b) (emphasis added). This effectively grants plain­tiffs the power not only to initiate the judicial process but also to hale their opponents into court. To be sure, given the role of the court clerk in issu­ing the summons, the latter power technically amounts to a Hohfeldian claim right on the part of the plaintiff to have the state exer­cise its power to hale people into court. 54 Cf. Herstein, How Tort Law Empowers, supra note 51, at 121 (“[Tort] victims hold a claim against the court to exercise its own legal power over tortfeasors in accordance with the victims’ rights for redress.”). But that claim right is essen­tially absolute, for the clerk has no discretion to withhold the summons so long as the plaintiff’s complaint satisfies the necessary formalities. 55 See 4A Charles Alan Wright et al., Federal Practice and Procedure § 1084, at 614 & n.5 (4th ed. 2015). Rather than require a clerk’s signature, some States even allow a plaintiff’s lawyer to sign the summons. See id. at 614–15 & n.6. In substance, then, if not in form, the Federal Rules assign the power to hale others into court to private parties rather than state officials.

That power is coercive. 56 See Barbara Allen Babcock et al., Civil Procedure: Cases and Problems 104 (4th ed. 2009) (“[B]ehind that innocent-looking piece of paper titled ‘Summons’ stands the full coercive power of the State.”). Once the summons issues and the plaintiff properly attempts to serve it on the defendant, the defendant must enter an appearance in the action, on pain of default. 57 See Fed. R. Civ. P. 55(a). The Rules thus empower a plaintiff to order other parties to appear in court and then to compel the court to coercively enforce that order—all virtually on the plaintiff’s own say-so. 58 The unilateral nature of this power has led at least one scholar to argue, on various doctrinal grounds, that Rule 4 is unconstitutional. See E. Donald Elliott, Twombly in Context: Why Federal Rule of Civil Procedure 4(b) Is Unconstitutional, 64 Fla. L. Rev. 895, 911–57 (2012). The effects of that coercion for the defendant can be significant, including the expense of hiring a lawyer and exposure to public scrutiny, as well as more formal legal consequences such as pre­judgment attach­ment of property. 59 See Fed. R. Civ. P. 64. On the other hand, the coercion may be relatively ephemeral; a spurious complaint can be dismissed fairly quickly, 60 See Fed. R. Civ. P. 12. and even a default or default judgment can, in certain circum­stances, be set aside. 61 See Fed. R. Civ. P. 55(c). The point is that the named defendant must generally appear before the court to get the complaint dismissed or the default or default judgment set aside, a burden imposed by the unilat­eral, unsupervised actions of the plaintiff. 62 The liberal rules of permissive joinder and third-party impleader only augment the plaintiff’s prosecutorial power, see Fed. R. Civ. P. 13, 14, 18, 20—especially considering how much latitude plaintiffs enjoy in deciding whom to join, see Stephen C. Yeazell, The Misunderstood Consequences of Modern Civil Process, 1994 Wis. L. Rev. 631, 654 [hereinafter Yeazell, Misunderstood Consequences].

A number of considerations confirm that the coercive power to hale others into court is a delegated state power. For one, we tend to associate the power to compel individuals to appear before state bodies with the state itself. That association is partly a product of history. Throughout much of the nineteenth century, American courts had discretion to review the substance of a plaintiff’s pleadings before issuing a sum­mons. 63 See Elliott, supra note 58, at 914–21. One finds formal vestiges of this history in the Federal Rules: Though instigated at the plaintiff’s sole discretion, a summons issues in the name of the court. It is formally the court, not the plaintiff, that orders the defendant to appear. 64 See Fed. R. Civ. P. 4(a)(1)(F)–(G) (requiring that a summons “be signed by the clerk” and “bear the court’s seal”). And it is formally the court, not the plaintiff, that coercively enforces that order with a default judgment. 65 Specifically, when the defendant fails to appear and the plaintiff seeks to recover “a sum certain,” the court clerk “must” enter a default judgment if the plaintiff so requests. Fed. R. Civ. P. 55(b)(1) (emphasis added). By contrast, when the plaintiff seeks an amount that is not certain, or when the defendant appears after having failed to answer the complaint, the court has broader discretion to decide whether to enter a default judgment. See Fed. R. Civ. P. 55(b)(2). The Rules thus formally profess state authorship over the very same pow­ers that they functionally put in the hands of private parties, one of the hallmarks of a delegation of state power.

For another, most agree on the public nature of the criminal prose­cu­torial power—the power to decide who must answer for alleged criminal conduct and to compel those individuals to appear in court. 66 To be sure, private parties prosecuted criminal actions throughout most of English history. See John H. Langbein, Renée Lettow Lerner & Bruce P. Smith, History of the Common Law: The Development of Anglo-American Legal Institutions 29–35, 578–89 (2009). And while the United States introduced public prosecutors early in its history, the practice of private prosecution persisted in certain quarters well into the nineteenth century. See id. at 746–49. But for at least a century, the prevailing view has been that criminal prosecution is exclusively a state function. See Roger A. Fairfax, Jr., Delegation of the Criminal Prosecution Function to Private Actors, 43 U.C. Davis L. Rev. 411, 413 (2009) (“Most observers reasonably view criminal prosecution as a function to be performed exclusively by the state.”). While the criminal prosecutorial power may have more significant conse­quences for defendants than the civil prosecutorial power, 67 Most notably, criminal prosecution exposes defendants to potential criminal punishment, whereas civil prosecution exposes defendants only to potential civil remedies. both ulti­mately reduce to the power to coercively hale people into court to answer for alleged wrongdoing.

One might resist this analogy on the ground that civil litigation typi­cally redresses private wrongs, as opposed to the public wrongs redressed through the criminal law and certain regulatory-style civil actions. More specifically, even conceding that the state delegates power when it autho­rizes private parties to bring criminal prosecutions, as well as certain civil enforcement actions, in its stead, one might still deny that any delegation extends to ordinary lawsuits seeking to vindicate private law rights, such as tort and contract actions. Seth Davis’s “state action” theory of standing doctrine is a sophisticated example of this position. According to Davis, “[s]tanding involves a government power when it authorizes a litigant to act on a government’s behalf to vindicate a government interest.” 68 Davis, supra note 18, at 590; cf. Metzger, supra note 4, at 1456 (“[T]he central criterion for singling out particular private delegations for enhanced scrutiny is whether they authorize private entities to act on the government’s behalf, a factor usually established by assessing whether the requirements of agency are met.”). “Stand­ing does not,” by contrast, “delegate the public power to enforce the law where it satisfies the government’s obligation to provide remedies to redress personal wrongs and to keep government accounta­ble, even if the litigant’s action would vindicate a public right.” 69 Davis, supra note 18, at 591. But cf. Sean Farhang, The Litigation State: Public Regulation and Private Lawsuits in the U.S. 8 (2010) (“The standing statutorily conferred on private actors to enforce public regulatory laws in court effectively licenses them to wield the coercive instruments of state power.”).

It is important to emphasize that Davis focuses on only one aspect of the civil prosecutorial power. In using the term “standing,” Davis is refer­ring simply to the right of a specific plaintiff to bring a specific claim against a specific defendant. 70 See Davis, supra note 18, at 595. The notion of governmental authorization does indeed offer a plausible test for determining when the exercise of that right involves a delegation of state power.

But as we have seen, the right to initiate a lawsuit does not exhaust the civil prosecutorial power in the American civil justice system; the Federal Rules also effectively grant plaintiffs the power to hale defend­ants into court, on pain of default. This power, which Davis does not sep­arately consider, is equally present even in cases Davis deems “private,” in which private parties are not authorized to vindicate a governmental interest. One need not maintain that “private standing and government standing are synonymous” 71 Id. at 601. or “give[] short shrift to [standing’s] role in realizing rights to remedies” 72 Id. at 604. to recognize that all civil cases involve this limited (though still significant) delegation, even if a narrower subset of civil cases involve an additional delegation of the power to enforce the law on the government’s behalf.

Davis appears to want to cabin the delegation involved in the civil prosecutorial power because he believes that any delegation of state power entails a duty on the part of the private party exercising the dele­gated power to subordinate her own private interests to public values. 73 See id. at 601–04. Such a duty, Davis suggests, would be inappropriate in ordinary private law cases, in which plaintiffs should have wide latitude to pursue their own interests in seeking a remedy, without regard to public values. 74 See id.; cf. BeVier & Harrison, supra note 42, at 1785 (premising constitutional law’s state action requirement “on the thesis that private individuals are principals, entitled to act to pursue their own interests, whereas government decisionmakers are agents, whose function is to further the interests of the citizens”). Davis purports to derive this limitation from civil recourse theory, 75 See Davis, supra note 18, at 601–04. yet civil recourse theorists themselves use the language of delegation to describe the civil prosecutorial power. 76 See, e.g., Benjamin C. Zipursky, Torts and the Rule of Law, in Private Law and the Rule of Law, supra note 46, at 139, 148 (referring to “the power the state affords to private individual litigants” (emphasis added)). More to the point, as the next Part shows, civil procedure already requires plaintiffs to restrain their use of the civil prosecutorial power for the sake of certain public values, even when they are not authorized to act on the government’s behalf but are instead permitted to pursue their own interests. Unsurprisingly, these restrictions are significantly less demanding than the restrictions Davis would impose on plaintiffs standing in for the government as law enforcer; after all, the power to enforce the law is typically an even more solemn state power than the power to coercively hale people into court. But while the latter delegation may be less significant than the former, it remains a delegation nonetheless.

2. Discovery. — The Federal Rules do not cease delegating state power once the plaintiff has summoned the defendant to appear in court. On the contrary, the state continues to delegate significant coer­cive power to both parties throughout the most consequential stage of civil litigation: discovery. Civil procedure scholars have long recognized that discovery is carried out primarily by the parties and their lawyers rather than the court; 77 See, e.g., Arthur R. Miller, Simplified Pleading, Meaningful Days in Court, and Trials on the Merits: Reflections on the Deformation of Federal Procedure, 88 N.Y.U. L. Rev. 286, 353 (2013) [hereinafter Miller, Simplified Pleading]; Yeazell, Misunderstood Consequences, supra note 62, at 651. it makes sense to conceptualize this feature of civil litigation, too, as a delegation of state power.

The structure of that delegation mirrors the structure of the delega­tion involved in the civil prosecutorial power. Just as the Rules grant plaintiffs the power to order defendants to appear in court, so they grant both parties the power to order each other, as well as nonparties, to disclose information. As the Supreme Court observed less than a decade after the promulgation of the Rules, “[e]ither party may compel the other to disgorge whatever facts he has in his possession.” 78 Hickman v. Taylor, 329 U.S. 495, 507 (1947). What was then true only functionally became true formally in 1970, when Rule 34 was amended to eliminate the (oft-ignored) requirement of a court order for discovery requests; the amendment authorized a party to directly request from the other party any document or object within the scope of discovery specified in Rule 26. 79 Fed. R. Civ. P. 34(a). As the Advisory Committee explained, the point of the amendment was to have Rule 34 “operate extrajudicially, rather than by court order.” Fed. R. Civ. P. 34(a) advisory committee’s note to 1970 amendment. One party can also demand that the other answer up to twenty-five interrogatories without a court order. 80 Fed. R. Civ. P. 33. A party’s discovery powers extend over not only the other party but nonparties as well. Under Rule 45, a party can subpoena a nonparty to pro­duce any discoverable documents or objects in her possession. 81 Fed. R. Civ. P. 34(c) (“As provided in Rule 45, a nonparty may be compelled to produce documents and tangible things or to permit an inspection.”). A party can also subpoena “any person” (including the other party) to “com­pel[]” her attendance at a deposition. 82 Fed. R. Civ. P. 30(a)(1) (oral examination); Fed. R. Civ. P. 31(a)(1) (written ques­tions); see also Fed. R. Civ. P. 45(a)(1). A party must seek the court’s permission to con­duct more than ten depositions (unless the parties stipulate to more). See Fed. R. Civ. P. 30(a)(2)(A)(i). A subpoena technically issues in the name of the court but is drafted by the party or her lawyer and is signed either by the lawyer or the court clerk, who must leave the sub­poena “otherwise in blank.” 83 Fed. R. Civ. P. 45(a)(3). At a deposition, a party can compel the deponent to answer any question that does not require the disclosure of privileged information, notwithstanding any objections (which are merely “noted on the record”). 84 Fed. R. Civ. P. 30(c)(2). Although a deposition must technically be taken before an “officer,” that person is usually just a stenographer or notary public who lacks authority to rule on objections or adjudicate disputes between the parties. Fed. R. Civ. P. 28. Time and time again, the Federal Rules allow the parties to order others to disclose information without prior judicial approval. 85 To be sure, at various points, the Rules encourage the parties to cooperate in plan­ning and conducting discovery. See Effron, supra note 38, at 138–48. But such cooperation typically does not involve nonparties, and it occurs (if at all) in the shadow of the coercive power that the Rules delegate to each party.

As with the civil prosecutorial power, the Rules back the parties’ dis­covery powers with the coercive power of the state—and once again leave courts with little control over how that power is wielded in the first instance. If one party refuses to comply with another party’s discovery request, the party who made the request can move for a court order com­pelling disclosure on pain of sanctions. 86 Fed. R. Civ. P. 37(a)–(b). This forces the resistant party to either oppose the motion or turn over the requested information; she cannot simply sit back and ignore the other party’s demands, lest she be sanctioned for noncompliance. The parties enjoy even more direct con­trol over the state’s coercive power with respect to discovery involving nonparties. A subpoena issued by a party under Rule 45 (whether signed by the party’s lawyer or the clerk) is treated as an order of the court, 87 9A Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 2453, at 396 (3d ed. 2008) (“Although an attorney-issued subpoena is in a sense merely the command of the attorney who completes the form, noncompliance with a subpoena by the recipient is an act in defiance of a court order.”). with noncompliance punishable as contempt. 88 Fed. R. Civ. P. 45(g). Although a nonparty can move to quash a subpoena, 89 Fed. R. Civ. P. 45(d)(3). she must take that affirmative step or otherwise face coercive sanctions simply because of the issuing party’s unilat­eral actions.

We should be careful not to overstate the extent of the powers exer­cised by private parties during discovery. Those powers have undoubtedly waned since 1938, as “successive amendments to the Federal Rules ha[ve] impressed limits on the extent of discovery.” 90 Arthur R. Miller, From Conley to Twombly to Iqbal: A Double Play on the Federal Rules of Civil Procedure, 60 Duke L.J. 1, 52 (2010) [hereinafter Miller, Double Play]; see also Miller, Simplified Pleading, supra note 77, at 353–56 (discussing changes to the discovery rules in the late twentieth century). Rule 26, for instance, was amended in 2015 to prohibit discovery that is not “propor­tional to the needs of the case.” 91 Fed. R. Civ. P. 26(b)(1). The impact of the 2015 amendments to the discovery rules will depend on individual judges’ willingness to employ the largely discretionary tools they authorize. See Morris A. Ratner, Restraining Lawyers: From “Cases” to “Tasks,” 85 Fordham L. Rev. 2151, 2167 (2017) (discussing the potential impact of the amend­ments on judicial case management); see also Andrew S. Pollis, Busting up the Pretrial Industry, 85 Fordham L. Rev. 2097, 2115 (2017) (expressing concern that the amendments leave judges with too much discretion over discovery). Some scholars have criticized the amendments, and particularly the proportionality requirement, as “anti-plaintiff.” See, e.g., Patricia Hatamyar Moore, The Anti-Plaintiff Pending Amendments to the Federal Rules of Civil Procedure and the Pro-Defendant Composition of the Federal Rulemaking Committees, 83 U. Cin. L. Rev. 1083, 1085–86 (2015) (arguing that various amendments to the Federal Rules over the last few decades have disadvantaged plaintiffs). Notwithstanding the increasing limits on discovery, however, it remains the case that parties still enjoy signifi­cant power to compel others to produce information without much judicial involve­ment. 92 See Amalia D. Kessler, Our Inquisitorial Tradition: Equity Procedure, Due Process, and the Search for an Alternative to the Adversarial, 90 Cornell L. Rev. 1181, 1256 (2005) [hereinafter Kessler, Our Inquisitorial Tradition] (“[D]espite the widely recognized problems with party control of discovery, the attempts we have made to pro­vide greater court control have been fairly limited—authorizing judges to hold scheduling conferences and to appoint pretrial discovery masters, but leaving primary responsibility with the parties.”). Indeed, “from a comparative perspective [the dis­covery power in American civil litigation] is broader and deeper than the powers exercised by private lawyers in any other legal system.” 93 Sklansky & Yeazell, supra note 37, at 694. While private parties’ discovery powers may have been curbed, they have not been eliminated. 94 This Article considers the implications of increased “managerial judging” sepa­rately below. See infra section I.C.2.

But why understand these powers as delegated state powers? Perhaps as much as any other civil procedure scholar, Stephen Yeazell has pointed toward an answer, and his account is worth quoting at length:

Discovery empowers litigants. In a civil law system litigants can only attempt to persuade a judge to seek information from the parties or others. American law, on the other hand, allows the litigants to demand information and testimony and backs those demands with legal sanctions. From such a comparative standpoint, the power delegated to ordinary litigants by the American discovery system is astounding. Ordinary civil litigants receive state sanction to require documents, depose witnesses under oath, and more. The penalties for noncompliance are substantial, including not only consequences to a litigant within the lawsuit, but also extending to possible imprisonment or fines of non-party witness[es] who fail to comply. We have put in the hands of civil litigants powers that in many legal systems only state officials enjoy. 95 Stephen C. Yeazell, Socializing Law, Privatizing Law, Monopolizing Law, Accessing Law, 39 Loy. L.A. L. Rev. 691, 699 (2006); see also Sklansky & Yeazell, supra note 37, at 694–95 (similar). For an unfavorable comparison of the American system of lawyer-driven discovery with civil law systems, see John H. Langbein, The German Advantage in Civil Procedure, 52 U. Chi. L. Rev. 823, 823–24 (1985).

To wit, in most other Western liberal democracies, judges exercise the very same discovery powers that private parties exercise in our civil justice system. 96 And if not judges, then administrative agencies. See Paul D. Carrington, Renovating Discovery, 49 Ala. L. Rev. 51, 54 (1997) (“Private litigants do in America much of what is done in other industrial states by public officers working within an adminis­trative bureaucracy.”). There is also a long history, extending well into the nineteenth century, of judicially conducted, inquisitorial-style discovery in American courts of equity, a practice that formed the basis for the party-conducted discovery authorized by the Federal Rules. 97 See Kessler, Our Inquisitorial Tradition, supra note 92, at 1206. On the rise and fall of equity procedure in the United States during the nineteenth century, see generally Amalia D. Kessler, Inventing American Exceptionalism: The Origins of American Adversarial Legal Culture, 1800–1877 (2017). This all suggests an association between discovery powers and the state—an association that still has some resonance in our civil justice system, if one is to judge from all of the formal markers of state authority that pervade the discovery rules, such as the fact that an “attorney acts as an officer of the court in issuing and signing subpoenas.” 98 Fed. R. Civ. P. 45 advisory committee’s note to 1991 amendment.

Even within the contemporary American legal system, state officials and institutions—most notably, police officers 99 See Sklansky & Yeazell, supra note 37, at 694 (“In many respects the lawyers representing the parties in civil actions have in their hands powers remarkably like those exercised by police investigators in criminal matters.”). and administrative agen­cies 100 See Farhang, supra note 69, at 8 (“The American civil discovery process effectively confers upon private litigants and their lawyers the same investigatory powers as federal agencies to compel sworn testimony and to disgorge documents . . . .”). —exercise investigatory powers that are functionally indistin­guishable from the discovery powers exercised by private parties in civil litigation. 101 Cf. Fed. R. Civ. P. 45 advisory committee’s note to 1991 amendment (“The 1948 revision of Rule 45 put the attorney in a position similar to that of the administrative agency, as a public officer entitled to use the court’s contempt power to investigate facts in dispute.”). The investigatory powers are considered public in those contexts not just because they are wielded by state agents but because they func­tion as necessary adjuncts of an unquestionably public power: the power to enforce the law. Yet, as we have seen, private parties gen­erally exercise the (coercive aspect of the) state’s law-enforcement power in our civil justice system. 102 See supra section I.B.1. Given the close relationship between discovery powers and the power to enforce the law, it makes sense to view the former as a state power too. 103 Cf. Carrington, supra note 96, at 54 (“[D]iscovery is the American alternative to the administrative state. We have by means of Rules 26-37, and by their analogues in state law, privatized a great deal of our law enforcement . . . .”).

Finally, the Federal Rules themselves implicitly underscore the pub­lic nature of the discovery powers exercised by private parties during civil litigation. In addition to authorizing the parties to compel each other to disclose information, the Rules establish a limited number of mandatory “initial disclosures”—certain categories of information that the parties must reveal to each other, independent of any request. 104 Fed. R. Civ. P. 26(a). Even these “required” disclosures, however, can be waived by the parties. See Fed. R. Civ. P. 26(a)(1)(A). This highlights the fact that the state can choose either to directly compel the disclosure of information itself or to grant that power to private parties. When the state makes the latter choice, as it does for most discovery matters, it seems difficult to describe that as anything other than a delegation of state power.

3. Settlement. — The case for conceptualizing the settlement of an ongoing lawsuit as a delegation of state power seems fairly straightfor­ward. With few exceptions, the court plays virtually no role in the settle­ment of a civil case; the parties may “generally elect to settle disputes and to choose their settlement terms without judicial oversight or inter­ference.” 105 Shiffrin, Remedial Clauses, supra note 6, at 429; see also Sklansky & Yeazell, supra note 37, at 698 (“[A]n ‘active role for the trial court in approving the adequacy of a settlement’ remains today ‘the exceptional situation, not the general rule.’” (quoting United States v. City of Miami, 614 F.2d 1322, 1331 (5th Cir. 1980))). The main exception is Rule 23(e), which requires judicial approval of class action settlements. Fed. R. Civ. P. 23(e). Judicial approval is also required for settlements in shareholder derivative actions, Fed. R. Civ. P. 23.1(c), and settlements reached under certain statutes, such as consent dec­rees in antitrust cases, see 15 U.S.C. § 16(b)–(e) (2012). Yet settlements bind the parties, and it is commonly thought that the power to reach a binding resolution of a private dispute is a paradigmatic state power. 106 See Boddie v. Connecticut, 401 U.S. 371, 375 (1971) (recognizing “the State’s monopoly over techniques for binding conflict resolution”); Davis & Hershkoff, supra note 10, at 535.

One might think that, in contrast to haling a person into court or compelling her to disclose information, settling a case is too far removed from state-sanctioned coercion to involve the exercise of state power. This impression is bolstered by the default procedure for dismissing a settled lawsuit in federal court. After the parties privately agree to settle the case, the plaintiff voluntarily dismisses the action by filing either a notice of dismissal (if before the defendant’s answer or summary judg­ment motion) or a stipulation of dismissal signed by all the parties (if later)—but either way, without having to obtain a court order. 107 Fed. R. Civ. P. 41(a)(1)(A). Given this procedure, “settlements do not implicate the judiciary’s judgment” as to either liability or remedy. 108 Shiffrin, Remedial Clauses, supra note 6, at 430; see also Sklansky & Yeazell, supra note 37, at 697 (“Settlement . . . is a private, largely invisible, contractual phenomenon; the judge plays a role only as a potential facilitator of private agreement. . . . The overwhelming majority of settlements require only that the parties agree; the court is told only that the plaintiff is dismissing the case.”). The state in no way lends its imprimatur to the parties’ resolution of the dispute but rather acts as if the lawsuit had never been filed. Formally, then, most settlement agreements look like any other private contract, and unless we are willing to completely jettison the public–private distinction, 109 See supra note 47. merely forming a voluntary agreement does not involve the exercise of a delegated state power. 110 Cf. Metzger, supra note 4, at 1444 (“[T]he power to resolve disputes on a nonconsensual basis, even subject to later state court review, is one that government cannot delegate to private actors without preserving some opportunity for prior review.” (emphasis added)).

Notwithstanding the dismissal procedure laid out in the Rules, how­ever, courts often issue an order dismissing a settled case and incorporate the parties’ settlement agreement into the order—without necessarily hav­ing supervised the negotiation of the agreement or approved its terms. 111 One of the primary ways this occurs in federal court is under the Supreme Court’s decision in Kokkonen v. Guardian Life Insurance Co. of America, which held that a federal district court may exercise “ancillary jurisdiction” over a subsequent action to enforce a settlement agreement, notwithstanding the absence of federal question and diversity jurisdiction, so long as it entered an order dismissing the original, settled action while also “incorporating” or “retaining jurisdiction over” the agreement in its order. 511 U.S. 375, 381–82 (1994). For other ways in which courts blur the formal boundary between settlements and court judgments, including through practices such as consent decrees, see Resnik, Contract, supra note 6, at 628, 630–33, 637, 662–65. This practice transmutes the parties’ private resolution of their dispute into a formal court order subject to ongoing coercive enforce­ment by the state, while forgoing the judicial scrutiny that typically precedes such an order. In this respect, settlements that are incorporated into a court order are difficult to distinguish from the delegations of state power involved in a summons or party-issued subpoena.

Even settlements that are not memorialized in a court order gen­erally end up functioning in practice much more like formal judgments than ordinary contracts. A prevailing plaintiff usually enforces a court judgment against a recalcitrant defendant through separate procedures that leave state officials with little discretion to withhold relief. 112 In federal district court, the enforcement of money judgments is generally gov­erned by the law of the State in which the court sits. See Fed. R. Civ. P. 69(a). State law typically provides for enforcement through a combination of writ-like procedures and separate enforcement actions. See Herstein, How Tort Law Empowers, supra note 51, at 124–30. Federal court judgments compelling specific actions are enforced via clerk-issued writs. Fed. R. Civ. P. 70(c)–(d). A violated settle­ment agreement, by contrast, is usually enforced through a separate State law action for breach of contract. 113 See J.J. Prescott & Kathryn E. Spier, A Comprehensive Theory of Civil Settlement, 91 N.Y.U. L. Rev. 59, 65 (2016). So far, court judg­ments and settlement agreements seem to inhabit completely different remedial worlds. But the contract law principles that govern the enforcement of settlement agreements can diverge from those that govern the enforcement of ordinary contracts in crucial respects. Compared with ordinary contracts, settlement agreements require an especially “strong showing” of invalid­ity 114 15A C.J.S. Compromise & Settlement § 33 (2017); see also, e.g., Hallock v. State, 474 N.E.2d 1178, 1180–81 (N.Y. 1984) (“Stipulations of settlement are favored by the courts and not lightly cast aside.” (citing In re Galasso, 320 N.E.2d 618, 618 (N.Y. 1974))). and are more difficult to rescind. 115 See, e.g., Compromise & Settlement, supra note 114, § 67 (“[A] mistake that would authorize the setting aside of an ordinary contract will not in all cases justify the setting aside of a compromise agreement.”). Their terms are also construed more liberally. 116 Id. § 34. And as with a court judgment, a “court will not inquire into the merits or validity of the original claim” when enforcing a settlement agreement. 117 Id. § 77. Some juris­dictions even go so far as to formally treat settlement agreements as court judgments, according them the same preclusive effect. 118 Id. § 27. These and other departures from general contract principles leave courts with significantly less discretion to refuse to coercively enforce settlement agreements, versus ordinary contracts—much as state officials have little discretion to refuse to coercively enforce court judgments. Yet in the case of settlement, it is the parties themselves, not state officials, who have authoritatively resolved their dispute. 119 Taking this feature of settlement agreements to its logical extreme, the Supreme Court has offered limited approval of cognovit (or confession of judgment) notes, provisions in loan agreements that require the debtor to prospectively waive her right to notice and a hearing before a court may enter judgment in favor of the creditor upon default. See D.H. Overmyer Co. v. Frick Co., 405 U.S. 174, 187 (1972). Cognovit notes thus go even further than ordinary settlement agreements, forcing the court to enter a formal, binding judgment without any appearance by the debtor and without regard to the merits of the case.

Other kinds of settlement practices grant private parties even more direct control over the state’s coercive power. For example, under Rule 68, a defendant can make an “offer of judgment” to the plaintiff. 120 Fed. R. Civ. P. 68. If the plaintiff accepts the offer, “[t]he clerk must then enter judgment”; if the plaintiff refuses and ends up securing a judgment no more favorable than the spurned offer, then she is liable to the defendant for “the costs incurred after the offer was made.” 121 Fed. R. Civ. P. 68(a), (d). This constitutes a kind of con­tingent delegation to the defendant: Simply by making a settlement offer during the litigation, a defendant can force the court either to enter a binding judgment (in the event the offer is accepted) or to sanction the plaintiff (in the event the offer is rejected and the plaintiff’s actual award does not exceed the offer). Either way, the defendant is compelling the court to deploy the state’s coercive power without any opportunity for the court to evaluate the basis on which it is being deployed. 122 A similar account can be given of stipulations made by the parties during litigation. As scholars have recognized, a stipulation can be understood as a partial settlement, in the sense that it resolves some important aspect of the case but without terminating the litigation. See Prescott & Spier, supra note 113, at 62–64. As such, a stipulation can also be understood as a partial delegation, for it limits the grounds on which a court can deploy the coercive power of the state but without fully determining how that power is deployed.

C. Doubting Delegation

At various points, the previous section addressed a number of specific objections to conceiving of civil procedure’s prosecution, discovery, and settlement powers as delegated state powers. This section briefly considers two additional, more thoroughgoing objections: (1) that private parties are formally “subordinate” to courts and (2) that “mana­gerial judges” oversee parties’ litigation conduct to such an extent that none of that conduct can constitute a private exercise of delegated state power.

1. Party Subordinance. — Even assuming that the powers discussed in the previous section are public, one still might question whether the Rules really allow private parties to exercise those powers in ordinary civil litigation—whether, that is, the powers are actually delegated. Scott Dodson’s elaboration of what he calls the “party subordinance” thesis suggests a particularly potent challenge along these lines. According to that thesis, “parties’ attempts to alter otherwise applicable procedures . . . are wholly unenforceable absent some legal authorization for judicial enforcement. And even when the law allows parties to exercise litigation choices, courts retain largely unfettered discretion—cabined only by law—to disregard or override those choices.” 123 Dodson, supra note 14, at 6–7; see also Davis & Hershkoff, supra note 10, at 562–63 (identifying various procedural rules that require courts to override party preferences). Dodson does not con­sider in any significant detail the specific contexts addressed in the previous section—prosecution, discovery, and settlement 124 Dodson’s only significant discussion of discovery, for example, focuses narrowly on Rule 29, which concerns parties’ stipulations regarding discovery procedures. See Dodson, supra note 14, at 22. —but his account nonetheless suggests that, even in those contexts, we should view courts, rather than litigants, as exercising ultimate control over the state’s coer­cive power.

Dodson’s party-subordinance thesis purports to describe the formal rules governing civil litigation, 125 See, e.g., id. at 7, 19 (emphasizing parties’ lack of “formal control over litigation”). but it is difficult to justify this exclu­sively formal focus. Surely what matters is who controls the state’s coer­cive power as a practical matter, not who bears nominal responsibi­lity for its exer­cise. 126 Cf. Davis & Hershkoff, supra note 10, at 537–40 (arguing that delegations can be “planned” or “de facto”). And as the previous section showed, the rules governing prosecution, discovery, and settlement functionally “empower[] or allow[] party dominance” while denying courts “discretion to override lawful party choices” with respect to the initial exercise of coercive power 127 Dodson, supra note 14, at 19. —in contrast to the impression the party-subordinance thesis might create.

Nor does the existence of even substantial judicial discretion necess­arily preclude a delegation of state power. After all, public law scholars conceptualize the “privatization” of various governmental programs as a delegation of state power even though administrative agencies must formally authorize private entities to administer the programs and retain discretion to override their decisions, which are also often subject to judi­cial review. 128 See, e.g., Metzger, supra note 4, at 1377–400; cf. Nagareda, supra note 18, at 191–98 (arguing that Rule 23 delegates lawmaking authority to lawyers in class actions, even though the court must ultimately approve any class settlement). That is presumably because, notwithstanding the govern­ment’s supervisory authority, private parties are still exercising state power in the first instance and thus can potentially cause serious harm before the government has a chance to intervene. Private parties likewise exercise multiple state powers in the first instance during ordinary civil litigation. While we may think that the prospect of judicial intervention serves as an adequate check on any potential abuse of those powers, the risk of abuse still exists. The delegation framework developed in the previous section calls attention to such concerns, whereas a narrow focus on private parties’ formal “subordinance” to courts tends to occlude them.

2. Managerial Judges. — Another objection is more practical than conceptual. This Part has argued that parties enjoy wide latitude in exercising various coercive powers during civil litigation, often with no meaningful judicial involvement. But many will no doubt dispute this depiction of judges as relatively passive bystanders. According to a popu­lar critique of contemporary federal civil practice, the demise of the civil trial has not diminished the judicial role so much as transformed it. Rather than adjudicate cases, judges now perform a more “managerial” function, shepherding cases and litigants through the pretrial phase. 129 This critique is most prominently associated with Judith Resnik. See Resnik, Diffusing Disputes, supra note 6, at 2844–50; Judith Resnik, Failing Faith: Adjudicatory Procedure in Decline, 53 U. Chi. L. Rev. 494, 496 (1986); Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374, 374 (1982); Resnik, Privatization, supra note 6, at 1803–06; Resnik, Contract, supra note 6, at 610–19; Judith Resnik, Trial as Error, Jurisdiction as Injury: Transforming the Meaning of Article III, 113 Harv. L. Rev. 924, 924 (2000); see also Freer, supra note 26, at 1508–27; Steven S. Gensler, Judicial Case Management: Caught in the Crossfire, 60 Duke L.J. 669, 669 (2010); Miller, Double Play, supra note 90, at 54–61, 78–81; Miller, Simplified Pleading, supra note 77, at 292–93; Jonathan T. Molot, An Old Judicial Role for a New Litigation Era, 113 Yale L.J. 27, 27 (2003); Todd D. Peterson, Restoring Structural Checks on Judicial Power in the Era of Managerial Judging, 29 U.C. Davis L. Rev. 41, 41 (1995). But see Steven Baicker-McKee, Reconceptualizing Managerial Judges, 65 Am. U. L. Rev. 353, 353 (2015) (defending managerial judging). This judicial management takes a number of different forms, exempli­fied for critics by the (discretionary) Rule 16 pretrial conference, during which the judge can structure the discovery process and pretrial motions practice. 130 Fed. R. Civ. P. 16. Another source of judges’ management authority is the Civil Justice Reform Act of 1990, 28 U.S.C. §§ 471–482 (2012), which requires federal district courts to develop procedures to combat expense and delay. While judicial management does not extend to the prose­cution of a case—plaintiffs still decide which claims to press and whom to name as defendants—it has fundamentally reshaped discovery and settle­ment; what were once largely private, consensual processes now often unfold under the court’s watchful eye. 131 Indeed, one of the main criticisms of “managerial judging” is that judges can use their managerial authority to unduly pressure parties to forgo trials and settle their cases. See Resnik, Contract, supra note 6, at 638–45; see also Marc Galanter & Mia Cahill, “Most Cases Settle”: Judicial Promotion and Regulation of Settlements, 46 Stan. L. Rev. 1339, 1342–46 (1994) (discussing factors that influence judicial intervention in settlements). One might think, therefore, that courts retain too much control over the state’s coercive powers during civil litigation for those powers to be delegated to the parties. 132 Cf. Robert G. Bone, Who Decides? A Critical Look at Procedural Discretion, 28 Cardozo L. Rev. 1961, 1967–71 (2007) [hereinafter Bone, Who Decides?] (describing various forms of discretion judges enjoy under the Federal Rules, though not focusing on the Rules considered in this Part); Stephen N. Subrin & Thomas O. Main, Braking the Rules: Why State Courts Should Not Replicate Amendments to the Federal Rules of Civil Procedure, 67 Case W. Res. L. Rev. 501, 508–10 (2016) (similar). But see Bone, Who Decides?, supra, at 1995 n.137 (“[M]uch of what happens at the pretrial stage takes place outside the judge’s view[].”).

It is hard to quarrel with this picture of the “managerial judge” as a descriptive matter, but it is important to distinguish two different ideas that often go under the heading “judicial management,” each of which has different implications for the delegation thesis advanced in the pre­vious section. First, there is management as housekeeping—administrative measures that help to ensure the orderly progression of the pretrial pro­cess. For discovery, judges can set schedules for making various disclosures and raising various objections. 133 See Fed. R. Civ. P. 16(b)(3); Ratner, supra note 91, at 2153. As for settlement, because judges gener­ally have no say in the content of settlement agreements, 134 See supra note 105 and accompanying text. they are largely relegated to encouraging the parties to settle and facilitating negotiations between them. 135 See Sklansky & Yeazell, supra note 37, at 701. Rule 16 expressly lists “facilitating settlement” as one of the goals of the pretrial conference. Fed. R. Civ. P. 16(a)(5). None of these forms of judicial involvement significantly curtails the parties’ discretion in exercising coercive powers; the parties can still largely compel whatever discovery, or settle on whatever terms, they would like. At least the “housekeeping” aspect of the “managerial judges” thesis, then, is compatible with concep­tualizing the parties’ discov­ery and settlement powers as delegations of state power.

Second, there is management as supervision—judicial scrutiny of, and interference with, the parties’ exercise of power during discovery and settlement. Although judges generally do not set strict limits on the scope of discovery, 136 See, e.g., John S. Beckerman, Confronting Civil Discovery’s Fatal Flaws, 84 Minn. L. Rev. 505, 561–71 (2000); John L. Carroll, Proportionality in Discovery: A Cautionary Tale, 32 Campbell L. Rev. 455, 463 (2010). they sometimes do, especially in complex cases. 137 See Ratner, supra note 91, at 2153–54, 2164. The court’s pretrial order may “modify the extent of discovery.” Fed. R. Civ. P. 16(b)(3)(B)(ii). Judicial limits on the scope of discovery are by no means uncontroversial; some scholars have argued that such limits undermine the private enforcement of public regulatory regimes. See Carrington, supra note 96, at 60–62. Likewise, judges in some cases go beyond simply encouraging settlement and actively participate in negotiations, even advocating specific settlement terms. 138 See Bone, Who Decides?, supra note 132, at 1968 (noting that “a judge can choose from a diverse menu of options” regarding settlement promotion); Bert I. Huang, Trial by Preview, 113 Colum. L. Rev. 1323, 1368 (2013) (“[C]ourts . . . vary as to how much of a role the future trial judge is allowed to play in settlement negotiations before trial.”); Sklansky & Yeazell, supra note 37, at 699 (“The civil rules cast the judge in the role—should she wish to assume it—of participant and even persuader, convening the parties and suggesting various possibilities of settlement and creative approaches to bridging disagreements between the parties.”); Yeazell, Misunderstood Consequences, supra note 62, at 657 (describing revisions to Rule 16 as “further encourag[ing] judges” to “encourag[e] settlement” and noting the debate about the desirability of such a role). See generally Ellen E. Deason, Beyond “Managerial Judges”: Appropriate Roles in Settlement, 78 Ohio St. L.J. 73, 77–104 (2017) (canvassing the different ways in which judges can be involved in settlement). Again, these practices, while not uncommon, are not the norm; notwithstanding the fact that some judges “are vigorously presiding, sometimes as managers, sometimes as judges,” it is still the case that “lawyers’ procedural choices will draw only sporadic judicial scrutiny.” 139 Yeazell, Misunderstood Consequences, supra note 62, at 638, 647; cf. Ratner, supra note 91, at 115 tbl.1 (identifying “[v]ariation among judges and cases” as a significant limit on the effectiveness of managerial judging).

The default in ordinary civil litigation thus remains a regime of party-driven discovery and settlement, with only the prospect of more significant judicial involvement lurking in the background. And even that involve­ment does not eliminate the delegations described in the previous section. Rather, as the next Part argues, we should understand “mana­gerial judging,” in its more supervisory guises, as a set of strategies for policing civil procedure’s delegations when private parties abuse dele­gated state power.

II. Policing Civil Procedure’s Delegations

This Part shows that civil procedure, even as it pervasively delegates coercive state power to private parties, also contains numerous doctrines that enable courts to police its delegations for abuse. Public law scholars have long recognized the potential for delegations of state power to be abused and the corresponding need for the state to police them. As section II.A argues, civil procedure’s delegations are no less prone to abuse, though the abuse assumes a different form in civil litigation than in the public law contexts on which scholars have primarily focused.

Section II.B then collects and classifies myriad procedural doctrines that allow courts to police private parties’ exercises of delegated state power during ordinary civil litigation. The doctrines fall into three main categories: those that rescind a delegated power, those that withhold enforcement from an exercise of delegated power, and those that punish abuse of delegated power. Whereas other scholars have tended to treat these various doctrines as unrelated to one another, this Part aims to show that they perform the common function of allowing the state to distance itself from private exercises of delegated power that offend especially important public values.

A. Abusing Delegated Procedural Powers

Recall the structure of civil procedure’s delegation problem: At vari­ous points during ordinary civil litigation, the Federal Rules effectively grant private parties the power to coerce others but without requiring prior judicial approval for those exercises of coercion. That creates a significant risk that the state’s coercive power will be abused—we have learned as much from the extensive literature on the “privatization” of many of the government’s administrative functions. 140 See generally Government by Contract: Outsourcing and American Democracy (Jody Freeman & Martha Minow eds., 2009) (discussing the growth of government and the outsourcing of governmental functions to private entities); Paul R. Verkuil, Outsourcing Sovereignty: Why Privatization of Government Functions Threatens Democracy and What We Can Do About It (2007) (evaluating the role of private contractors in performing governmental functions); Metzger, supra note 4 (analyzing the constitutionality of the “privatization” of governmental functions); Minow, Public and Private Partnerships, supra note 1 (considering the effects of privatization on public values). But just as the delegations at issue in public law contexts differ from civil procedure’s delegations in important respects, so we need a different conception of abuse of state power to understand the unique risks posed by the delegations in ordinary civil litigation.

Public law’s conception of abuse of delegated state power begins from what Gillian Metzger has called “the principle of constitutional accounta­bility.” 141 Metzger, supra note 4, at 1373. According to that principle, “the Constitution imposes limits on the actions that governments can take,” limits that “apply to all exercises of government power, whether wielded by officially public or nominally private entities.” 142 Id. Any conception of abuse of delegated state power in civil litigation should start from the same general principle. But public law scholars have cashed out the notion of “accountability” in ways that do not capture the concerns raised by civil procedure’s delega­tions of state power.

Consider again Seth Davis’s conception of a delegation as authoriza­tion to act on the government’s behalf. 143 See supra notes 68–75 and accompanying text. From that conception, Davis derives a notion of accountability according to which “the Constitution requires all government action to be justified by reference to some public value.” 144 Davis, supra note 18, at 607 (internal quotation marks omitted) (quoting Cass R. Sunstein, Naked Preferences and the Constitution, 84 Colum. L. Rev. 1689, 1692 (1984)); see also id. at 615 (“Liberal democracies demand that government action be based upon public reasons, not raw preferences . . . .”). Other scholars take a similar view. See, e.g., Lawrence, supra note 42, at 659 (endorsing the principle that governmental officials should exercise their authority to promote “some conception of what is good for the community”). A private party, on this view, must affirmatively pursue some public-regarding purpose when exercising delegated state power for the delegation to be legitimate; if she instead employs the power to some pri­vate end, she violates constitutional norms and abuses the delegation. 145 Cf. Restatement (Second) of Agency § 387 (Am. Law Inst. 1958) (providing that an agent must “act solely for the benefit of the principal in all matters connected with his agency” (emphasis added)); Restatement (Third) of Agency §§ 8.01–.06 (Am. Law Inst. 2006) (same). This constitutional principle is embodied in the private nondelegation doctrine, which holds that the government violates due process when it delegates its lawmaking power in ways that allow private parties to regulate others’ conduct for private purposes. See Carter v. Carter Coal Co., 298 U.S. 238, 311 (1936). Davis draws on the private nondelegation doctrine in developing his state action theory of standing. See Davis, supra note 18, at 610–11, 615–16.

That conception of accountability (and abuse) makes sense insofar as the government has authorized private parties to act on its behalf. But as we have seen, the government can delegate state power to private parties without necessarily making them its agents or forbidding them to pursue their own interests. 146 Metzger appears to concede this point. See Metzger, supra note 4, at 1468–69 (acknowledging the possibility that the state can delegate power to private parties without creating an agency relationship). Sometimes, the government grants private parties power not so they can directly pursue public objectives in its stead but so they can more effectively pursue private ends that society deems valuable. 147 To be sure, by using delegated state power to pursue their own ends, private parties can, through an invisible-hand-like mechanism, also further public goals—even in traditional private law cases. Cf. William J. Novak, The Myth of the “Weak” American State, 113 Am. Hist. Rev. 752, 769–71 (2008) (identifying delegation as a state-building strategy). See generally Farhang, supra note 69 (examining how the state uses private enforcement to augment its regulatory capacities). But the existence of such incidental effects does not convert private parties into the government’s agents. These private exercises of power are nevertheless delegations because of the nature of the power private parties wield—in the context of civil litigation, the power to coerce others to take actions in relation to the binding resolution of private disputes. 148 Perhaps the clearest doctrinal support for this conception of a delegation is the “civil Batson” case. See Edmonson v. Leesville Concrete Co., 500 U.S. 614 (1991). There, the Supreme Court held that peremptory challenges constitute a delegation of state power, see id. at 626–28, notwithstanding the fact that litigants may legitimately exercise those challenges for private purposes, see id. at 626 (“Though the motive of a peremptory challenge may be to protect a private interest, the objective of jury selection proceedings is to determine representation on a governmental body.”). For further discussion of Edmonson as a delegation-policing strategy, see infra section II.B.2.

This different conception of delegation calls for different concep­tions of accountability and abuse. When a private party is delegated state power but is not required to affirmatively pursue some public purpose, she remains accountable by not offending certain norms governing the exercise of state power in the course of pursuing her private ends. Con­versely, a private party abuses delegated state power when she violates those norms. The worry in civil procedure is not that private parties will fail to further public goals—they generally are not required to do so—but that they will “commandeer[] the judicial process” and “force a publicly subsidized court to sacrifice public benefits for purely private interests.” 149 Dodson, supra note 14, at 15 (emphasis added). The risk is one of harm to public values rather than neglect of public duties.

Thus articulated, this account of abuse of civil procedure’s delega­tions remains largely formal rather than substantive. It identifies the normative structure of the abuse as the violation of important public values, without elaborating on the specific content of those values. It is beyond the scope of this Article to provide a comprehensive account of the pub­lic values whose violation constitutes an abuse of delegated state power, 150 I plan to develop a more complete account in future work. but those values have many sources. One is the Federal Rules themselves. The notion that the Rules are value-free 151 See, e.g., Miller, Double Play, supra note 90, at 90 (noting the rulemakers’ “desire to keep the original Federal Rules textually simple and value neutral as much as possible”). has long since been debunked; as scholars have recognized, the Rules embody a number of important values, including justice, accuracy, transparency, and effi­ciency. 152 See, e.g., Miller, Simplified Pleading, supra note 77, at 288 (“When the Federal Rules were promulgated in 1938, they embodied a justice-seeking ethos.”); Stephen N. Subrin, How Equity Conquered Common Law: The Federal Rules of Civil Procedure in Historical Perspective, 135 U. Pa. L. Rev. 909, 943–48 (1987) (arguing that the values of “equity lived on through the Federal Rules”). Beyond the Rules, substantive law reflects other fundamental normative commitments of our society, such as nondiscrimi­nation. These and other values are, to varying degrees, implicated in private parties’ exercise of delegated state power during civil litigation.

Many procedural rules and doctrines seek to safeguard these values by requiring courts to honor them in the course of adjudicating disputes. Thus, courts must follow certain procedures before rendering decisions, judges may not discriminate against parties based on race, court proceed­ings must generally be open to the public, and so on. But because private parties also wield state power during civil litigation, they, too, can violate the norms governing the exercise of that power. Hence the need for delegation-policing doctrines—doctrines that allow the state to specifically address exercises of delegated state power that offend especially impor­tant public values.

B. Civil Procedure’s Delegation-Policing Doctrines

Civil procedure generally affords private parties wide leeway to exer­cise delegated state power during civil litigation. Rather than hold private parties to the same standards that govern courts and other public institu­tions, civil procedure brooks a significant amount of abuse of delegated state power as one of the costs of having an adversarial, as opposed to inquisitorial, system. 153 Cf. Daniel Markovits, Arbitration’s Arbitrage: Social Solidarity at the Nexus of Adjudication and Contract, 59 DePaul L. Rev. 431, 445–48 (2010) (noting that litigants and their lawyers may act contrary to various public values). There comes a point, however, at which civil pro­cedure deems private parties’ abuse of delegated power too egregious to tolerate and allows the state to counteract that abuse in various ways. This section examines multiple doctrines that police the outer limits of private parties’ discretion to wield delegated state power during civil litigation.

At the outset, it is useful to distinguish doctrines that incidentally prevent abuse of delegated state power by limiting parties’ opportunities to exercise it. One particularly controversial example is pleading. By tightening the general pleading standard under Rule 8 of the Federal Rules, 154 See Ashcroft v. Iqbal, 556 U.S. 662, 678–79 (2009); Bell Atl. Corp. v. Twombly, 550 U.S. 544, 562–63 (2007). the Supreme Court attempted to make it more difficult for com­plaints to survive a motion to dismiss, thereby restricting plaintiffs’ access to discovery—and thus their opportunities to abuse their discovery-related powers. 155 This Article brackets the empirical debate over whether Twombly and Iqbal have in fact had this effect. For a recent overview of, and contribution to, that debate, see generally Jonah B. Gelbach, Material Facts in the Debate over Twombly and Iqbal, 68 Stan. L. Rev. 369 (2016) (presenting empirical evidence regarding the adjudication of defendant-filed summary judgment motions after Twombly and Iqbal). A number of subject-matter-specific rules and statutes impose similar limits on how plaintiffs may prosecute certain kinds of claims. 156 E.g., Private Securities Litigation Reform Act of 1995, Pub. L. No. 104-67, § 104, 109 Stat. 737, 757 (1995) (imposing special procedural requirements for private actions under the Securities and Exchange Act of 1934); Fed. R. Civ. P. 9(b) (imposing a heightened pleading requirement for fraud claims). See generally Jessica Erickson, Heightened Procedure, 102 Iowa L. Rev. 61 (2016) (describing heightened procedural requirements for securities class actions, medical malpractice lawsuits, and product liability cases). State anti-SLAPP laws are another example. As exemplified by California’s statute, anti-SLAPP laws apply to lawsuits that implicate the defendant’s free-speech rights, such as libel actions. See Cal. Civ. Proc. Code § 425.16(b), (e) (2018). Defendants in such actions may file “a special motion to strike,” which stays discovery and requires the plaintiff to show, based on whatever evidence she already has, that she is more likely than not to prevail on her claim. See id. § 425.16(b), (g). Such doctrines do not directly regulate the grounds on which parties may exercise the powers delegated to them during civil litigation but rather seek to reduce the overall costs of litigation by restricting par­ties’ access to those powers. The doctrines nonetheless have the effect of limiting parties’ opportunities to exercise—and thus to abuse—delegated powers. One can therefore understand the doctrines as embodying a kind of prophylactic strategy for reducing the risk of abuse of delegated state power before it has occurred.

That kind of strategy has significant shortcomings. Most notably, the state faces considerable decision costs in trying to anticipate and forestall every potential form of abuse of civil procedure’s delegated powers. Any rule that seeks to restrict access to those powers will consequently be overbroad, which can undermine other values, such as access to justice. 157 This is a major basis for criticism of Twombly and Iqbal. See, e.g., A. Benjamin Spencer, Pleading and Access to Civil Justice: A Response to Twiqbal Apologists, 60 UCLA L. Rev. 1710, 1732–33 (2013) (arguing that “plausibility pleading is overinclusive and hope­lessly subjective,” that it restricts litigants’ access to justice, and that it is a misguided attempt to curb discovery abuse). At the same time, even overbroad rules will let some instances of abuse slip by, putting the state in the untenable position of having to tolerate even egregious forms of abuse.

Perhaps because of such drawbacks, civil procedure does not primar­ily rely on this ex ante strategy of limiting access to delegated state power. Rather, it generally affords private parties broad access to its delega­tions 158 Cf. Richard Marcus, “American Exceptionalism” in Goals for Civil Litigation, in Goals of Civil Justice and Civil Procedure in Contemporary Judicial Systems 123, 123, 134–35 (Alan Uzelac ed., 2014) (arguing that American civil procedure is “exceptional” in, among other things, “relaxing burdens on plaintiffs” and permitting “extremely broad discovery”). and then employs an alternative, ex post strategy whereby the state reviews a party’s exercise of delegated power for its conformity with public values and, if the abuse is too much to bear, interferes with or disavows the private party’s actions, making clear that her abuse of state power cannot stand unchallenged. 159 To be clear, civil procedure’s delegation-policing doctrines are enshrined in various rules, statutes, and court decisions before they are applied in any case, and so are in that sense “ex ante.” But they are “ex post” in the specific sense that courts apply them in response to an abusive exercise of delegated power, whereas access-restricting doctrines, such as pleading rules, attempt to deny private parties the opportunity to exercise delegated power before any abuse has occurred. The ex post policing function resembles the “anti-opportunism” or “safety-valve” function that Henry Smith has ascribed to various equitable doctrines. See Henry E. Smith, Equity as Second-Order Law: The Problem of Opportunism 19 (Harvard Pub. Law Working Paper No. 15-13, 2015),
abstract=2617413 (on file with the Columbia Law Review); see also Samuel L. Bray, The System of Equitable Remedies, 63 UCLA L. Rev. 530, 578–84 (2016) (describing various “equitable constraints” that help to counteract party abuse of equitable remedies).
Civil procedure is replete with doctrines that perform precisely this delegation-policing function. 160 This Article’s distinction between access-restricting and delegation-policing doc­trines differs from Joanna Schwartz’s distinction between “gateway” and “pathway” rules in civil procedure. See Joanna C. Schwartz, Gateways and Pathways in Civil Procedure, 60 UCLA L. Rev. 1652, 1654–55 (2013). According to Schwartz, “[g]ateway rules limit cost, delay, and unjust outcomes by preventing cases from proceeding to the next stage of litiga­tion, while pathway rules limit cost, delay, and unjust outcomes by controlling troublesome aspects of litigation while allowing cases to proceed.” Id. at 1659. Although most delegation-policing doctrines are pathway rules in this sense, a few terminate the litigation and thus qualify as gateway rules. See, e.g., infra note 179 and accompanying text (discussing sua sponte dismissals). Not all pathway rules, moreover, concern stages of the litigation that involve the exercise of delegated state power. And even among those pathway rules that do, some operate automatically before any abuse has occurred, whereas the delegation-policing doctrines operate through the exercise of judicial discretion after the fact. See, e.g., Fed. R. Civ. P. 26(a)(1)(A) (requiring the parties to make certain “initial disclosures” “without awaiting a discovery request” or court order); Schwartz, supra, at 1659 (classifying Rule 26(a) as a pathway rule). Nor do civil procedure’s delegation-policing doctrines constitute a form of “ad hoc procedure.” Pamela K. Bookman & David L. Noll, Ad Hoc Procedure, 92 N.Y.U. L. Rev. 767, 772–73 (2017) (“Ad hoc procedure is designed to address a procedural problem that arises in a pending case or litigation. It is then applied retroactively to that pending case or litigation in order to achieve a desired result.”). Although those doctrines operate through the exercise of case-by-case judicial discretion, they are developed before any dispute has arisen rather than during the litigation. See supra note 159.

Before this section turns to those doctrines, two caveats are in order. First, this Article makes no claim about the origins of civil procedure’s delegation-policing doctrines. It argues only that various procedural doctrines currently enable courts to police civil procedure’s delegations for abuse, whether or not they were intentionally developed for that purpose. 161 Cf. Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 415 (1996) (arguing that various First Amendment doctrines smoke out illicit governmental purposes, regardless of whether they were intentionally devised to perform that function).

Second, to classify the doctrines considered in this section as dele­gation-policing doctrines is not necessarily to defend them against those who might consider them an inadequate response to the various forms of abuse that can occur during civil litigation. Indeed, civil procedure does not police its delegations consistently but instead tends to address only the most egregious forms of abuse of certain delegations, while ignoring the abuse of other delegations altogether. 162 For example, there is no meaningful check on abuse of Rule 68’s offer-of-judgment procedure. See supra notes 120–121 and accompanying text; see also infra notes 299–300 and accompanying text (discussing criticisms of discovery). That said, civil procedure’s delegation-policing efforts, however imperfect, address a significant amount of abuse of delegated state power in ordinary civil litigation—especially when compared with the permissiveness of the current arbitra­tion regime. 163 See infra section IV.A.1. Civil procedure also responds to significantly more abuse than does public law, which generally ignores “bad faith” conduct by governmental actors. See David E. Pozen, Constitutional Bad Faith, 129 Harv. L. Rev. 885, 918 (2016) (arguing that “judges have frequently ignored constitutional bad faith”).

Civil procedure’s delegation-policing doctrines can be classified into three main categories, according to how intrusively they interfere with private parties’ exercise of delegated state power. The following typology is not necessarily comprehensive, nor are the boundaries separating its different categories perfectly clean. The point is to identify the most pro­minent types of delegation-policing strategies in civil procedure.

1. Rescinding Delegations. — The most intrusive delegation-policing doctrines effectively rescind a delegation when it is being abused, and then either reassign responsibility for exercising the delegated state power at issue to a state official or, in extreme cases, deny the private party responsible for the abuse all further access to that power. Among the rescissory doctrines that substitute a state official for the parties are some of the practices that civil procedure scholars have dubbed “mana­gerial judging.” 164 See supra section I.C.2. For example, the Federal Rules permit courts to limit the scope of discovery beyond Rule 26’s capacious relevance standard. 165 See Fed. R. Civ. P. 16(b)(3)(B)(iii); Fed. R. Civ. P. 26(b)(1). A court can, of course, exercise this discretion for many different rea­sons, including simply to save time and expense. But at least when a court limits discovery in response to the parties’ abusive discovery practices, one can understand the court’s action as (partially) rescinding the parties’ power to decide what discovery to compel and lodging that power back with a state official—namely, the judge. 166 The rescission is only partial because the parties still retain broad discretion to compel discovery within the limits set by the court.

The effect is similar when a court appoints a special master to oversee discovery. 167 See Fed. R. Civ. P. 53 (authorizing the court to appoint a special master without the parties’ consent when there is “some exceptional condition”); Acad. of Court-Appointed Masters, Appointing Special Masters and Other Judicial Adjuncts: A Handbook for Judges and Lawyers § 2.1, at 17–18 (2d ed. 2009) (identifying as one “exceptional condition” warranting appointment of a master the fact that “[t]here will be unusual discovery or evidentiary problems requiring continued oversight”). Such oversight can be quite extensive, including supervising depositions, 168 See Acad. of Court-Appointed Masters, supra note 167, § 1.2, at 4 (“Sometimes a discovery master will sit in on depositions that are particularly contentious.”). during which the Rules otherwise give the par­ties free rein. While courts will often appoint a discovery master for rea­sons of efficiency, such as handling complex technical issues involved in electronic discovery, 169 See id. § 1.3, at 4. they sometimes also use special masters to curb discovery abuse. In such cases, the court rescinds some of the coercive discovery powers delegated by the Rules, transferring the parties’ discre­tion for exercising those powers to a state official. 170 Although a private individual can be appointed as a special master, she is still a state official in the sense that she’s appointed by and accountable to the court. But see Kessler, Our Inquisitorial Tradition, supra note 92, at 1238–60 (arguing that, in many cases, special masters are vulnerable to capture by the parties and thus not truly independent).

Courts can also effectively retract the parties’ power to reach a bind­ing resolution of their dispute through a settlement agreement. Recall that judges generally have no say over settlements; with few exceptions, the parties can settle their dispute on whatever terms they’d like without the court’s approval. 171 See supra note 105 and accompanying text. In practice, however, some judges go beyond merely encouraging settlement and facilitating negotiations and shape the actual substance of a settlement agreement. 172 See supra note 138 and accompanying text. Although the parties formally retain control (it is the parties, after all, who decide whether to stipulate to a dismissal), in some cases the court can exert such influence that it assumes a kind of adjudicatory authority as a practical matter. Inso­far as the court exercises that de facto authority to resist what it regards as unfair or otherwise abusive settlement terms, it partially rescinds the parties’ power to authoritatively resolve their dispute themselves. 173 To the extent class action settlements involve a delegation of state power, see Nagareda, supra note 18, at 191–98, courts can similarly rescind that delegation by aggressively exercising their authority under Rule 23(e) to review and approve such settlements. But see, e.g., William B. Rubenstein, The Fairness Hearing: Adversarial and Regulatory Approaches, 53 UCLA L. Rev. 1435, 1443–46 (2006) (arguing that judges exercise insufficient oversight of class action settlements). For critical accounts of the role judges play in policing non-class “aggregate settlements,” see generally Howard M. Erichson, The Role of the Judge in Non-Class Settlements, 90 Wash. U. L. Rev. 1015 (2013); David M. Jaros & Adam S. Zimmerman, Judging Aggregate Settlement, 94 Wash. U. L. Rev. 545 (2017). But see Andrew Bradt & D. Theodore Rave, The Information-Forcing Role of the Judge in Multidistrict Litigation, 105 Calif. L. Rev. 1259, 1259 (2017) (urging judges to perform an “information-forcing” function in non-class aggregate settlements); Linda S. Mullenix, Policing MDL Non-Class Settlements: Empowering Judges Through the All Writs Act 1–2 (Pub. Law & Legal Theory Research Paper Series, Paper No. 572, 2017), (on file with the Columbia Law Review) (defending judicial involvement in non-class aggregate settlements under the All Writs Act).

Many civil procedure scholars—including many of the proponents of the delegation critique—criticize this kind of judicial involvement in settlement negotiations for, among other things, potentially pressuring weaker parties to relinquish their right to their day in court. 174 See supra note 131; see also Bone, Who Decides?, supra note 132, at 2012–14 (identifying other potential problems with judicial involvement in settlement negotiations). That is a legitimate concern, but viewing such involvement as a delegation-policing strategy shows it to be much less pernicious, and even salutary. Given that most parties are going to settle with or without the court’s approval, we should welcome judicial involvement in ordinary settle­ments as a way for the state to reassert control over its coercive adjudica­tory power when that power stands to be abused by the private parties to whom it has been delegated.

Beyond these transsubstantive delegation-rescinding strategies, vari­ous subject-matter-specific procedural rules can similarly be understood as ways of reasserting state control over delegated powers, particularly the prosecutorial power, in cases of abuse. Consider the provision of the False Claims Act (FCA) that allows the federal government to intervene in and take over the prosecution of a qui tam action. 175 See 31 U.S.C. § 3730(b) (2012). At first glance, that provision seems to simply reflect the fact that, in a qui tam action, the federal government is the real party in interest, while the private relator is a mere stand-in seeking to vindicate the government’s interests. Given this structure, the FCA’s intervention provision appears to allow the government to prosecute fraud directly in cases in which private enforcement proves inadequate. And that is indeed the provision’s pri­mary purpose, but not its sole one. The FCA makes clear that the govern­ment does not automatically displace the relator when it intervenes; the relator retains a (minimal) stake even after the government has inter­vened and taken over the action. 176 See, e.g., id. § 3730(c)(1) (providing that the relator generally “shall have the right to continue as a party to the action”); id. § 3730(c)(2)(A) (providing that a qui tam action cannot be dismissed at the government’s behest without first giving the relator an opportunity to be heard); id. § 3730(c)(2)(B) (imposing a similar hearing requirement for settlements); id. § 3730(c)(3) (permitting the relator to proceed with the action even if the government declines to do so). The statute further recognizes that the relator’s stake includes a degree of prosecutorial control over the action and that the relator can abuse that control by exercising it for improper purposes. Most notably, the statute authorizes the court to limit, at the government’s request, the relator’s participation in the action in various ways—not just to avoid unnecessary expense and pro­mote effi­ciency but also when the relator is acting for “purposes of harassment.” 177 Id. § 3730(c)(2)(C)–r(D). This suggests another (if secondary) purpose served by government intervention in qui tam actions: reclaiming the delegated prosecutorial power for state officials in cases of abuse. 178 For empirical evidence regarding Department of Justice (DOJ) intervention practices that is potentially consistent with this hypothesis, see generally David Freeman Engstrom, Public Regulation of Private Enforcement: Empirical Analysis of DOJ Oversight of Qui Tam Litigation Under the False Claims Act, 107 Nw. U. L. Rev. 1689 (2013). Engstrom identifies the FCA’s intervention provision as an agency litigation gatekeeping strategy that is “affirmative,” “binding,” and “retail,” with “active displacement” and “strong control rights.” Engstrom, Litigation Gatekeepers, supra note 48, at 646 tbl.2. For other variations on this policing strategy, see id. at 650–54.

Each of the foregoing delegation-policing strategies rescinds the delegated power at issue and substitutes a state official for the private party who abused that power. But the state can also rescind a delegation without substitution, terminating the party’s action rather than allowing the action to continue under government supervision. For example, federal courts may dismiss a complaint sua sponte, without waiting for the defendant to appear and file a motion to dismiss under Federal Rule of Civil Procedure 12—though only when the claims are patently frivo­lous and only after affording the plaintiff certain procedural protections, such as notice, an opportunity to be heard, and leave to amend her complaint. 179 See 5B Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 1357, at 409 & n.4 (3d ed. 2004). A sua sponte dismissal effectively rescinds the delegated power to coercively hale others into court, thereby preempting a plain­tiff’s (perceived) abuse of that power before she has managed to inflict any serious harm.

Similarly, but more controversially, the Prison Litigation Reform Act authorizes federal courts to dismiss sua sponte an action brought by a prisoner challenging prison conditions if the action, among other things, “is frivolous, malicious, [or] fails to state a claim upon which relief can be granted.” 180 42 U.S.C. § 1997e(c)(1) (2012); see also 28 U.S.C. § 1915A(b)(1) (2012). What is more, under the so-called three-strikes rule, a prisoner generally may not be granted in forma pauperis status if, on three or more prior occasions, he filed an action or appeal that was dis­missed on any of those grounds. 181 28 U.S.C. § 1915(g). Given the limited financial resources of most prisoners, that is typically tantamount to permanently revoking a prisoner’s power to prosecute claims in federal court. One need not countenance the application of this rule to prisoner litigation in parti­cular in order to recognize it as an example of a more general strategy that, along with other rescissory doctrines, allows the state to respond to ongoing abuse of delegated procedural powers.

2. Withholding Enforcement. — Of course, rescinding a delegated state power works as a delegation-policing strategy only insofar as a private party’s abuse of that power is not yet complete. It does little good to rescind the summonsing power after the plaintiff has haled the defendant into court, or the subpoena power after a party has issued a subpoena, or the settlement power after the parties have stipulated to a dismissal. In each of those cases, one private party has already exercised the state’s coercive power over another. So rather than seek to undo that exercise of coercion (which is impossible), the state must try to limit its effects. A number of doctrines aim to do exactly that by allowing the state to avoid facilitating, or giving further effect to, an abusive exercise of delegated power.

Unsurprisingly, this strategy does not figure prominently in the prosecution context. Because a summons issues automatically once the plaintiff presents it to the court clerk, 182 See Fed. R. Civ. P. 4(b). the state cannot refuse to enforce the plaintiff’s exercise of the summonsing power; it has little opportunity to spare the defendant from having to appear and answer the complaint, on pain of default. The best the state can do in most cases is to dismiss the complaint after the defendant has appeared and thereby deny the plaintiff access to additional state powers, particularly those relating to discovery. 183 It can also punish the plaintiff’s exercise of the prosecutorial power with sanctions, a strategy considered below. See infra section II.B.3.

The state can, by contrast, more effectively avoid facilitating a party’s abuse of the various discovery powers. If a party makes an abusive discov­ery request and, after the other party refuses, moves to compel, 184 See Fed. R. Civ. P. 37(a). the court cannot undo the coercion the moving party has already exercised. It can, however, limit the effects of that coercion by denying the motion to compel. The opposing party still must affirmatively resist the request and oppose the motion (or preemptively move for a protective order before the motion has been filed), but at least she won’t have to actually turn over the requested information. The same is true when a court grants a motion to quash a subpoena. 185 See Fed. R. Civ. P. 45(d)(3). By refusing in each case to lend the state’s support to a party’s abusive exercise of delegated coercive power during discovery, the court avoids compounding the abuse. 186 Similarly, even as it delegates the state’s power to reach binding resolutions of private disputes, see supra note 111, the Supreme Court’s decision in Kokkonen v. Guardian Life Insurance Co. of America, 511 U.S. 375 (1994), at least theoretically gives federal courts a kind of de facto review power over settlement agreements that they would otherwise lack. For if the parties are not completely diverse and want their agreement to be enforceable in federal court, they will have to persuade the court either to retain jurisdiction over the agreement or to incorporate it into the order dismissing the case. That effectively empowers the court to condition federal jurisdiction over a future enforcement action on a fair agreement and, conversely, to refuse to lend support to an agreement it deems unfair by refusing jurisdiction. Although proponents of the delegation critique of arbitration have been somewhat wary of Kokkonen, see, e.g., Resnik, Contract, supra note 6, at 627, the case creates space for a delegation-policing strategy in an area where parties otherwise enjoy nearly unfettered discretion.

While one finds examples of this delegation-policing strategy at the core of the Federal Rules, it is even more prominent in other parts of the procedure canon. Perhaps the clearest example is Edmonson v. Leesville Concrete Co., in which the Supreme Court extended the constitutional ban on race-based peremptory challenges in the selection of criminal juries to civil cases 187 500 U.S. 614, 618–28 (1991). —so-called civil Batson. 188 Batson v. Kentucky, 476 U.S. 79 (1986) (establishing a constitutional ban on race-based peremptory challenges in criminal cases). Because the constitutional guarantee of equal protection applies only to “state action,” 189 Edmonson, 500 U.S. at 619. the key issue in Edmonson was whether a private litigant (or her lawyer) could be considered a state actor in exercising peremptory challenges against prospective jurors. The Court applied a two-pronged test to resolve that question and found the first prong—whether a peremptory challenge is “a right or privilege having its source in state authority” 190 Id. at 620. —to be easily met, given that peremptory challenges are available only insofar as the government chooses to authorize them. 191 See id. at 620–21.

As for the second prong—“whether a private litigant in all fairness must be deemed a government actor in the use of peremptory chal­lenges”—the Court considered the extent to which the peremptory challenge is imbued with public authority. 192 Id. at 621. Peremptory challenges, the Court noted, depend on “the overt, significant participation of the government,” from the government’s general administration of the jury system to the court’s involvement in voir dire. 193 Id. at 622–24. But more relevant for this Article’s argument, the Court also emphasized that litigants exercise peremptory challenges in the course of discharging “a traditional func­tion of the government”: selecting the members of the jury, a govern­mental body. 194 Id. at 624. The peremptory challenge, in other words, is one aspect of a “delegated” state power. 195 See id. at 626–28. And the Court clearly saw litigants as abusing that delegation when they exercise peremptory challenges in a racially discriminatory manner.

The Court’s refusal to enforce race-based peremptory challenges can accordingly be seen as a delegation-policing strategy akin to the strategy available to courts under the discovery rules. Like a discovery request, a peremptory challenge ultimately depends on court enforce­ment to be fully effective. 196 See id. at 624 (noting that “[w]ithout the direct and indispensable participation of the judge, who beyond all question is a state actor, the peremptory challenge system would serve no purpose”). As the Court put it in Edmonson, in exercising a peremp­tory challenge, a litigant “invokes the formal authority of the court, which must discharge the prospective juror.” 197 Id. Edmonson retained the baseline rule that courts will automatically dis­charge a prospective juror in response to a peremptory challenge, thus preserving the delegation of state power involved in such challenges. But the decision also created room for courts to withhold the state’s support from abusive exercises of that delegated power by refusing to discharge a prospective juror in response to a race-based peremptory challenge. Although the state cannot undo the harm inherent in a discriminatory peremptory challenge, it can avoid facilitating that discrimination and thereby limit the extent of a party’s abuse of delegated state power.

A similar delegation-policing strategy emerges from a series of Supreme Court decisions over the latter third of the twentieth century concerning the constitutionality of State prejudgment attachment and replevin statutes. In those decisions, the Court repeatedly invalidated laws that allowed a creditor–plaintiff to invoke state assistance in seizing property possessed by the debtor–defendant based on the plaintiff’s conclusory assertion of an interest in the property. 198 See Connecticut v. Doehr, 501 U.S. 1, 18 (1991) (invalidating a Connecticut statute that allowed a plaintiff to attach real estate owned by the defendant during a pending lawsuit simply upon filing a statement of “probable cause” and posting a bond, without first affording the defendant notice or an opportunity to be heard); N. Ga. Finishing, Inc. v. Di-Chem, Inc., 419 U.S. 601, 606–07 (1975) (invalidating a Georgia statute that allowed a plaintiff to garnish the defendant’s assets during a pending lawsuit simply upon filing a conclusory affidavit and posting a bond, without first affording the defendant an opportunity to be heard); Fuentes v. Shevin, 407 U.S. 67, 96 (1972) (invalidating Florida and Pennsylvania summary replevin statutes that allowed any private party to seize personal property that she alleged to be wrongfully held by the possessor simply upon the posting of a bond, without first affording the possessor any notice or an opportunity to be heard); Sniadach v. Family Fin. Corp. of Bay View, 395 U.S. 337, 341–42 (1969) (invalidating a Wisconsin garnishment statute that allowed a creditor to temporarily freeze the wages of the debtor pending the outcome of the garnishment action simply by serving the summons, without first affording the debtor any notice or an opportunity to be heard). Although the formal legal question in each case was whether the lack of an adversarial predeprivation hearing violated procedural due process, 199 See Doehr, 501 U.S. at 9; Di-Chem, 419 U.S. at 603; Fuentes, 407 U.S. at 80; Sniadach, 395 U.S. at 338. the Court could not straightforwardly apply its standard due process analysis, since the government had not initiated the seizure. The Court instead had to consider the permissibility of “enabling an individual to enlist the aid of the State to deprive another of his or her property by means of the prejudgment attachment or similar procedure.” 200 Doehr, 501 U.S. at 9. This was essentially a question of the propriety of delegating state power.

More specifically, each of the statutes invalidated by the Court resembled the delegations considered in Part I, in that a private party could get the state to deploy its coercive power for her own purposes more or less on her own say-so and without meaningful judicial review. As the Court explained in one of the decisions, “[t]he statutes . . . abdicate effective state control over state power. Private parties, serving their own private advantage, may unilaterally invoke state power to replevy goods from another.” 201 Fuentes, 407 U.S. at 93. This structure mirrors that of the delegations involved in a summons or discovery subpoena. 202 To be sure, the Court later made clear that a statute does not delegate state power simply by licensing self-help. See Flagg Bros., Inc. v. Brooks, 436 U.S. 149, 157 (1978) (noting that the “total absence of overt official involvement plainly distinguishes this case from earlier decisions imposing procedural restrictions on creditors’ remedies”); cf. Andrew S. Gold, Private Rights and Private Wrongs, 115 Mich. L. Rev. 1071, 1074–80 (2017) (arguing that the law sometimes allows self-help, and in such cases, the person engaging in self-help is enforcing her own rights, not exercising state power). But a subsequent decision held that there is a delegation of state power “whenever officers of the State act jointly with a creditor in securing the property in dispute,” which is the case “when the State has created a system whereby state officials will attach property on the ex parte application of one party to a private dispute.” Lugar v. Edmondson Oil Co., 457 U.S. 922, 932–33, 942 (1982) (construing the phrase “under color of state law” in 42 U.S.C. § 1983 but treating that issue as more or less synonymous with the constitutional state action inquiry); cf. Zoë Sinel, De-Ciphering Self-Help, 67 U. Toronto L.J. 31, 58 n.70 (2017) (arguing that “replevin is not self-help”).

In all but one of the cases in which the Court found a delegation of state power, it deemed the delegation impermissible and invalidated the statute at issue because the State had failed to maintain adequate control over its coercive power. 203 See, e.g., Fuentes, 407 U.S. at 91–94. Those decisions can thus be understood as prohibiting delegations of state power when there were insufficient checks on how private parties exercised that power.

The Court instead pursued a delegation-policing strategy in the one decision in which it both found state action under a repossession statute and upheld the statute as consistent with due process. In Mitchell v. W.T. Grant Co., the Court sustained a Louisiana statute that allowed a creditor, on ex parte application to a judge, to provisionally sequester personal property held by the debtor subject to a vendor’s lien. 204 416 U.S. 600, 618–19 (1974). The Court distinguished its prior decisions invalidating similar statutes on multiple grounds but emphasized in particular that the predeprivation hearing, though ex parte, was before a judge rather than a court clerk or other ministerial official, and that the creditor had to submit a sworn affidavit setting out the specific factual allegations entitling her to possession of the property. 205 See id. at 607–10, 616–17; see also N. Ga. Finishing, Inc. v. Di-Chem, 419 U.S. 601, 607 (1975) (distinguishing a Georgia garnishment statute from the statute at issue in Mitchell on the ground that it required only a conclusory affidavit to be filed with the clerk). The Court also noted that the questions to be resolved by the judge concerned “uncomplicated matters that lend themselves to documentary proof.” Mitchell, 416 U.S. at 609. Notwithstanding these differences, it still makes sense to view the Louisiana statute as delegating state power. After all, the predep­rivation judicial review provided by the statute, while perhaps more than “a mere ministerial act,” 206 Mitchell, 416 U.S. at 616 n.12. did not amount to that much more. “Since the procedure . . . is completely ex parte,” Justice Stewart explained in dissent, the judge “can do little more than determine the formal suffi­ciency of the plaintiff’s allegations before ordering the state agents to take the goods from the defendant’s possession.” 207 Id. at 629, 632 (Stewart, J., dissenting). The provision for judicial review thus did not eliminate the delegation conferred by the other parts of the statute; the baseline remained virtually automatic enforcement of a creditor’s sequestration request. And yet, the statute did allow the state to police that delegation to some degree, by refusing to lend its support to patently abusive sequestration requests. That limited opportunity for the state to withhold enforcement may well have been insufficient to address all or even most abuse, but it is nevertheless an example of a kind of delegation-policing strategy employed in other parts of civil procedure.

3. Punishing Abuse. — Besides rescinding a delegation of state power when it is being abused or withholding enforcement to limit the effects of the abuse, the state can punish the private party who commits the abuse with various sanctions after the fact. This delegation-policing strategy tends to be less intrusive than the first two, as it does not attempt to actually interfere with private parties’ exercise of delegated state power; parties can still exercise the delegated power as they will, though they must suffer the consequences for any abuse they perpetrate. Nor, of course, can sanctions undo the abuse. 208 In very rare circumstances, civil procedure allows courts to undo an abusive exercise of delegated state power. For example, when a party obtains a final judgment in her favor by fraud or other misconduct, the court may annul the judgment under Federal Rule of Civil Procedure 60(b)(3). Yet they do allow the state to dis­avow an abusive exercise of delegated state power by expressing condem­nation of the private party’s conduct.

The paradigmatic example of this delegation-policing strategy is Rule 11 of the Federal Rules. Rule 11 authorizes courts to impose sanc­tions on a party or her lawyer for, among other things, “presenting” a written submission “for any improper purpose, such as to harass, cause unneces­sary delay, or needlessly increase the cost of litigation.” 209 Fed. R. Civ. P. 11(b)(1). To be sure, “presenting” written submissions to the court encompasses a broad swath of litigation conduct, including routine adversarial actions that in no way involve an exercise of delegated state power. A party does not, for instance, exercise delegated state power by filing a motion for a prelimi­nary injunction or for summary judgment. Filing a written submission does, however, frequently involve an exercise of delegated state power—most notably, when a plaintiff files a complaint and thereby triggers the issuance of a summons.

In punishing parties for taking such actions for an “improper purpose,” Rule 11 reflects the conception of abuse of delegated state power developed in the previous section. Rule 11, like the rest of civil procedure, grants private parties significant latitude to use delegated state power to pursue private ends; a private purpose is not necessarily an “improper purpose,” and so does not necessarily render an exercise of delegated state power abusive. At the same time, however, private parties must observe certain public norms in wielding state power, and Rule 11 imposes liability for offending those norms. This allows the state to repu­diate, if not reverse, private parties’ abusive exercises of delegated state power. 210 Among the sanctions a court may impose under Rule 11 are nonmonetary sanc­tions and monetary penalties payable either to the court or to the other party. See Fed. R. Civ. P. 11(c)(4). When a court imposes a nonmonetary remedy or a penalty payable to the court, the sanction is purely punitive; when it imposes a penalty payable to the other party (such as an award of costs or attorney’s fees), the sanction is also compensatory. Especially when it’s too late to rescind the delegation or with­hold enforcement, punishing abuse may be the only practical way for the state to maintain a degree of accountability for the procedural powers it has delegated to private parties. 211 Scholars have criticized Rule 11 on various grounds over the decades. Among other criticisms, they have faulted courts for imposing sanctions inconsistently across different categories of parties and have worried that the risk of sanctions deters potential plaintiffs from bringing meritorious lawsuits. See Stephen B. Burbank, The Transformation of American Civil Procedure: The Example of Rule 11, 137 U. Pa. L. Rev. 1925, 1937 (1989); Melissa L. Nelken, Sanctions Under Amended Federal Rule 11—Some “Chilling” Problems in the Struggle Between Compensation and Punishment, 74 Geo. L.J. 1313, 1339–43 (1986); William W. Schwarzer, Rule 11 Revisited, 101 Harv. L. Rev. 1013, 1015–17 (1988). Rule 11 was significantly revised in 1993, largely to address these kinds of concerns. See Carl Tobias, The 1993 Revision of Federal Rule 11, 70 Ind. L.J. 171, 171 (1995). Interestingly, the normative valence of Rule 11 seems to have shifted in recent years, as some scholars now invoke Rule 11 as a preferable alternative to other measures aimed at curbing abusive litigation practices, such as heightened pleading standards. See Spencer, supra note 157, at 1722, 1732–33.

Civil procedure employs similarly punitive delegation-policing strat­egies throughout civil litigation. While Rule 11 expressly does not apply to discovery, 212 See Fed. R. Civ. P. 11(d). Rule 26 authorizes courts to sanction a party for any discovery request that is “interposed for any improper purpose, such as to harass, cause unnecessary delay, or needlessly increase the cost of liti­gation.” 213 Fed. R. Civ. P. 26(g)(1)(B)(ii), (3). Rule 45 likewise allows courts to sanction a party for issuing and serving a subpoena that “impos[es] undue burden or expense.” Fed. R. Civ. P. 45(d)(1). And pursuant to their inherent powers, federal courts may order a party to pay the other side’s attorney’s fees as a sanction for “bad-faith” litigation conduct. 214 See Chambers v. NASCO, Inc., 501 U.S. 32, 50 (1991). At least where these sanctions are imposed in connection with an exercise of delegated state power, they constitute an important delegation-policing strategy.

The state can sanction abusive exercises of delegated state power not only during the litigation but also through collateral proceedings. Fore­most among the collateral mechanisms are the tort law causes of action for malicious prosecution and abuse of process. Like Rule 11, these torts embody the notion of abuse developed in the previous section, in that each imposes tort liability for litigation conduct taken for some “impro­per purpose” or with “malice”—instituting a lawsuit in the case of mali­cious prosecution and, depending on the jurisdiction, a wide range of other procedural actions in the case of abuse of process. 215 See 3 Dan B. Dobbs, Paul T. Hayden & Ellen M. Bublick, The Law of Torts § 592, at 407–15 (2d ed. 2011); id. § 594, at 418–28. The action­able abuse, once again, consists in transgressing certain public values in the course of exercising delegated state power rather than merely using such power to pursue some private objective. While it is relatively difficult to prevail on these causes of action, 216 For example, in addition to having to prove that the abusive lawsuit was instituted for an improper purpose, a malicious-prosecution plaintiff must prove that the suit was not supported by probable cause and was terminated in her favor. See id. § 592, at 407–08. A substantial minority of jurisdictions also require malicious-prosecution plaintiffs to show that they suffered some kind of “special injury” from the abusive lawsuit. See id. § 593, at 415–18. they attach at least some negative consequences to abusive exercises of delegated state power, especially when sanctions during the litigation are unavailable or prove insufficient.

To be clear, actions for malicious prosecution or abuse of process are instituted by other private parties, not state officials. So insofar as they con­stitute a punitive delegation-policing strategy, that strategy is only indirect; the delegation-policing function is itself delegated to private parties. This shows that, in contrast to Rule 11 sanctions, the malicious-prosecution and abuse-of-process torts are primarily compensatory and only secondarily puni­tive. Nevertheless, in affording victims of abusive exercises of dele­gated state power some kind of remedy, the state imposes limits on civil procedure’s various delegations.

Finally, courts can sanction lawyers as well as the parties they rep­resent, both during the litigation and through collateral proceedings. During the litigation, Rule 11 and Rule 26 authorize a court to sanction a party’s lawyer instead of, or in addition to, the party herself. 217 See Fed. R. Civ. P. 11(c)(1), 26(g)(1)(3). Other provisions are directed exclusively at lawyers, such as a federal statute allowing courts to hold a lawyer partially liable for the opposing side’s attorney’s fees and costs when she “multiplies the proceedings in any case unreasonably and vexatiously.” 218 28 U.S.C. § 1927 (2012). Such conduct often involves abusing some of the delegated state powers involved in ordinary civil litigation. 219 See, e.g., Apex Oil Co. v. Belcher Co. of N.Y., 855 F.2d 1009, 1019–20 (2d Cir. 1988) (holding that § 1927 applies to discovery abuse). As for collateral mechanisms, a number of professional ethics rules cover lawyer conduct that can constitute an abuse of delegated state power. 220 See, e.g., Model Rules of Prof’l Conduct rs. 1.2(d), 3.1, 3.3, 3.4, 3.5(d) (Am. Bar Ass’n 2016). One of the subsidiary functions of the system of lawyer discipline, it turns out, is policing some of civil procedure’s delegations on behalf of the state. Sanctioning lawyers in all of these various ways makes perfect sense as a delegation-policing strategy insofar as a party has subdelegated a dele­gated state power to her lawyer and the lawyer is at least partially responsible for abusing that power. As with sanctions directed at the par­ties, the point is to target the individual who actually abused delegated state power so that the state can distance itself from her misconduct.

III. Delegation Policing as Complicity Avoidance

Considering civil procedure’s various delegation-policing doctrines raises the question why the state should police the delegated powers in ordinary civil litigation for abuse. At first blush, the answer seems obvious: Especially because state power is coercive power, abuse of delegated state power can cause significant harm to others; the various delegation-policing strategies help to ameliorate, if not prevent, that harm. 221 Cf. Davis, supra note 18, at 615–21 (arguing that abuse of the delegated power to enforce the law on the government’s behalf can harm third parties in various ways); Metzger, supra note 4, at 1462–63 (arguing that abuse of the delegated power to administer govern­mental programs on the government’s behalf can unfairly deprive third parties of access to government benefits). This concern is surely an important part of the story. Indeed, it helps to explain why, from a delegation-policing perspective, many of the doctrines discussed in the previous Part are overbroad, in that they regu­late not just exercises of delegated state power but adversarial litigation conduct more generally. Those doctrines reflect a generic concern with the harms that can be inflicted through civil litigation, whether by abuse of delegated state power or otherwise.

Some of the other doctrines, however, have a narrower focus, uniquely targeting exercises of delegated state power, and then only certain kinds of abuse of that power. This Part identifies an additional reason, beyond general harm prevention, for the state to focus on some forms of procedural abuse to the exclusion of others. It argues that the various delegation-policing doctrines reflect not only an other-regarding concern with limiting the harm private parties can suffer from the abuse of delegated state power, but also a self-regarding concern on the part of the state with distancing itself from private wrongdoing that involves the violation of certain public values, so as to preserve its own moral integrity and that of the political community it serves. This Article labels the latter concern “complicity avoidance.” 222 This concern differs from more familiar anxieties about procedural private ordering’s effects on judicial “integrity” or “legitimacy”—the worry that private parties’ preferences will displace judges’ independent judgment. See Bone, Party Rulemaking, supra note 38, at 1384–97; Davis & Hershkoff, supra note 10, at 547–51; Dodge, supra note 10, at 764–70; cf. Shiffrin, Remedial Clauses, supra note 6, at 419–25 (criticizing certain kinds of remedial clauses in contracts on similar grounds).

As with this Article’s account of abuse of delegated state power, 223 See supra section II.A. its account of complicity avoidance is primarily formal rather than substan­tive. This Part seeks to elucidate the general normative structure of the state’s role in facilitating private wrongdoing during civil litigation, not to comprehensively catalog the types of private wrongdoing that threaten to implicate the state and the political community or to specify all the circumstances in which the state has an obligation to extricate itself from those misdeeds.

Section III.A introduces the concept of complicity avoidance, reveal­ing its logic to be immanent in certain corners of civil procedure doctrine, particularly civil Batson. Section III.B then shows that state complicity in private wrongdoing is a problem that recurs throughout the law. It considers, in particular, two instances of complicity avoidance beyond civil procedure: the version of the state action doctrine applied in Shelley v. Kraemer and Seana Shiffrin’s revisionist account of the unconscionabi­lity doctrine in contract law. Section III.C defends complicity avoidance as a distinct normative logic that cannot be reduced to the dominant norma­tive logic in civil procedure, procedural due process. Finally, section III.D grounds complicity avoidance in liberal political theory, according to which the state claims a monopoly on the legitimate use of coercive power and exercises that power in the name of the political community. These features of political power raise the moral stakes of delegation, threatening to implicate all of us in private wrongdoing.

A. Complicity Avoidance in Civil Procedure

In decisions applying civil procedure’s various delegation-policing doctrines, courts rarely discuss in significant detail the values served by policing delegations for abuse. But when they do, they do not focus exclu­­sively on the harm suffered by victims of abusive exercises of delegated state power; they also speak of the abuse’s “tainting” or “corrupting” the civil justice system, or the wrongdoer’s making the court “a party to” the abuse. 224 See infra notes 237–238 and accompanying text. Courts also worry, in other words, about the state’s becoming complicit in the abuse.

This distinct, self-regarding concern is perhaps most explicit in Edmonson, the civil Batson case. To be sure, the Court acknowledged the personal “injury to excluded jurors” from discriminatory peremptory chal­l­enges. 225 Edmonson v. Leesville Concrete Co., 500 U.S. 614, 628 (1991). But also at stake, in the Court’s view, was the state’s own moral integrity. “By enforcing a discriminatory peremptory challenge,” the Court explained, a “court ‘has not only made itself a party to the [biased act], but has elected to place its power, property and prestige behind the [alleged] discrimination,’” and thus “in a significant way has involved itself with invidious discrimination.” 226 Id. at 624 (alterations in original) (quoting Burton v. Wilmington Parking Auth., 365 U.S. 715, 725 (1961)); see also id. at 628 (“[R]acial discrimination in the qualification or selection of jurors offends the dignity of persons and the integrity of the courts.” (internal quotation marks omitted) (quoting Powers v. Ohio, 499 U.S. 400, 402 (1991))). This complicity in private discrimination “mars the integrity of the judicial system and prevents the idea of democratic government from becoming a reality.” 227 Id. at 628. Edmonson essentially imposed an obligation on all courts to avoid complicity by refusing to give effect to discriminatory peremptory challenges—that is, by withholding state support from abusive exercises of a delegated state power.

In her dissent in Edmonson, Justice O’Connor rejected the majority’s broad notion of state responsibility for private wrongdoing in civil litiga­tion. The civil trial, as she saw it, “is by design largely a stage on which private parties may act”; 228 Id. at 632 (O’Connor, J., dissenting). But see id. at 633–34 (“The peremptory [challenge] is, by design, an enclave of private action in a government-managed proceeding.”). the state merely “erects the platform” and “does not thereby become responsible for all that occurs on it.” 229 Id. at 632; see also id. at 641 (maintaining that “the performance of a lawyer’s duties in a courtroom” does not constitute state action). For this alternative conception of the civil trial, Justice O’Connor relied pri­marily on the Court’s decision in Polk County v. Dodson, which had held that a public defender does not act “under color of state law” within the meaning of 42 U.S.C. § 1983 when executing a lawyer’s traditional “adver­sarial functions.” 230 454 U.S. 312, 320 (1981). A lawyer, the Dodson Court reasoned, performs what is “essentially a private function”: “advancing ‘the undivided interests of his client.’” 231 Id. at 318–19 (quoting Ferri v. Ackerman, 444 U.S. 193, 204 (1979)). The Edmonson majority acknowledged that not every act by a party or her lawyer during civil litigation necessarily constitutes state action. 232 See Edmonson, 500 U.S. at 627–68 (explaining that in most civil cases, “the initial decision whether to sue at all, the selection of counsel, and any number of ensuing tactical choices in the course of discovery and trial may be without the requisite governmental character to be deemed state action”). Nonetheless, it viewed the peremptory challenge as not just another “private function,” but an exercise of delegated state power that threatened to implicate the state in the grievous moral wrong of racial discrimination. 233 Id.

In fairness, Justice O’Connor had a point when she observed that the peremptory challenge is really not so different from other “adver­sarial act[s]” taken during civil litigation. 234 Id. at 642 (O’Connor, J., dissenting). To take two examples, sum­monses and subpoenas seem to share the features that, in the Edmonson Court’s view, render a peremptory challenge state action: They similarly depend on “the overt, significant participation of the government,” 235 Id. at 622 (majority opinion). in that each is issued in the name of the court and is backed by the state’s coercive power, and neither seems to be any less of a “traditional func­tion of the government” 236 Id. at 624. than a peremptory challenge.

Perhaps unsurprisingly, then, courts employ the logic of complicity avoidance beyond the state action context and in cases involving proce­dural powers other than peremptory challenges. For example, in decisions imposing sanctions under Rule 11 and the discovery rules, courts often describe the abuse of prosecution and discovery-related powers as “taint­ing” or “corrupting” the justice system or making the court “a party to” or even “complicit in” private wrongdoing, and they invoke sanctions as a way of removing that taint. 237 See, e.g., Xyngular Corp. v. Schenkel, 200 F. Supp. 3d 1273, 1328 (D. Utah 2016) (“Anything short of dismissal will fail to remove the taint that Schenkel has caused and fail to adequately deter future litigants from engaging in this type of self-help discovery.”); see also Mark v. New Orleans City, No. 15-7103, 2017 WL 2152367, at *6 (E.D. La. May 16, 2017) (acknowledging “cases where courts have . . . noted that the legitimacy of courts is injured when courts are ‘complicit in allowing these shakedown schemes to continue’” but ultimately declining to impose sanctions under Rule 11 (quoting Doran v. Del Taco, Inc., 373 F. Supp. 2d 1028, 1031 (C.D. Cal. 2005))). Similar rhetoric appears in abuse-of-process and malicious-prosecution decisions. 238 See Cordes v. Outdoor Living Ctr., Inc., 781 S.W.2d 31, 34 (Ark. 1989) (likening abuse of process to extortion or coercion, and noting the possibility that the process might be “tainted by a willful act” of one of the parties); Antur Realty Corp. v. Rivera, 442 N.Y.S.2d 732, 733 (Civ. Ct. 1981) (arguing that an “abuse of process” can “make th[e] court a party to . . . a tainted cause”); Wolfe v. Little, No. 18718, 2001 WL 427408, at *2 (Ohio Ct. App. Apr. 27, 2001) (explaining that the abuse-of-process tort “developed to provide a remedy in situations where an appropriate legal procedure has been . . . corrupted in order to accomplish some ulterior motive for which a court proceeding was not intended”); see also Kolka v. Jones, 71 N.W. 558, 565–66 (N.D. 1897) (“If the law will not punish such conduct . . . the courts themselves will seem to have forsaken their high function as protectors and vindicators of invaded rights, and to have become, instead, the accomplices of evil men.”). Whether or not the exercise of these other procedural powers formally constitutes state action as a mat­ter of constitutional law, courts view the abuse of those powers as breach­ing the public–private divide and staining the state’s moral integrity.

The notion of complicity in these decisions does not depend on the narrower conception of a delegation articulated by public law scholars such as Davis and Metzger. 239 See supra notes 68–75, 143–145 and accompanying text. The state can, in other words, become com­plicit in an abusive exercise of delegated state power even though the private party wielding that power isn’t acting on the government’s behalf or constrained to pursue its purposes. In Edmonson, for instance, the Court perceived a threat of complicity, notwithstanding the fact that pri­vate par­ties may legitimately exercise peremptory challenges for their own ends. 240 See Edmonson, 500 U.S. at 626 (“Though the motive of a peremptory challenge may be to protect a private interest, the objective of jury selection proceedings is to determine representation on a governmental body.”). The risk of complicity arises not because private parties are stand­ing in for the state but because they are exercising its coercive powers. This means that the state faces a pervasive risk of complicity in civil litiga­tion. Because a private party need not be authorized to act as the state’s agent in order to render it complicit in her wrongdoing, every exercise of delegated state power in civil litigation potentially implicates the state’s moral integrity. And because there is a fine line between a legitimate pri­vate purpose and an improper one—an aggressive discovery request, for example, can quickly shade into a vexatious one—a significant number of those exer­cises of delegated power will in fact prove to be abusive, and thus, for the state, corrupting.

And yet, civil procedure’s delegation-policing doctrines only mitigate the threat of complicity rather than eliminate it. They do so to varying degrees. The delegation-rescinding doctrines afford the state the most protection, deeming the risk of complicity too great to allow private parties to continue to exercise state power unsupervised by state officials. The doctrines that withhold enforcement from certain exercises of dele­gated power, by contrast, are somewhat more permissive: They allow pri­vate parties to continue to exercise—and thus to abuse—delegated state powers but refuse to lend the state’s support to exercises of those powers that offend especially important public values. Finally, the doctrines that punish abusive exercises of delegated state power tolerate the most complicity, censuring only the worst forms of abuse after they have occurred. Given the limits of the various delegation-policing strategies, the state ends up incurring a good deal of complicity in abusive exercises of delegated power that can be neither thwarted nor repudiated. This is by no means to deny the significant benefits of delegating procedural powers to private parties. 241 These benefits are discussed further at infra section III.D.2. But we should also recognize that those bene­fits come at a considerable moral cost to the state. 242 In this respect, civil procedure provides private parties with a moral subsidy—not just the financial subsidy that other scholars have identified. See, e.g., Judith Resnik, Money Matters: Judicial Market Interventions Creating Subsidies and Awarding Fees and Costs in Individual and Aggregate Litigation, 148 U. Pa. L. Rev. 2119, 2131–37 (2000) (identifying various taxpayer-financed subsidies for civil litigants); Stephen J. Ware, Is Adjudication a Public Good? “Overcrowded Courts” and the Private Sector Alternative of Arbitration, 14 Cardozo J. Conflict Resol. 899, 900 (2013) (criticizing the civil justice system for squander­ing “public resources that would be better spent on parties who deserve the subsidy”). I plan to explore this idea further in future work.

B. Complicity Avoidance Beyond Civil Procedure

Civil procedure’s delegation-policing doctrines represent a distinctly procedural response to a distinctly procedural problem. Ordinary civil litigation delegates a number of procedural powers to private parties. In abusing those powers, private parties can render the state complicit in their wrongdoing. The delegation-policing doctrines, in their different ways, allow the state to avoid that complicity by distancing itself from pri­vate parties’ exercises of delegated procedural powers.

Sometimes, however, the threat of complicity arises from the substance of a party’s legal claims. Even when private parties exercise delegated procedural powers scrupulously, they can still jeopardize the state’s moral integrity if they are using civil litigation to seek the state’s support for some objectionable goal and the substantive law entitles them to relief. This different iteration of the complicity problem calls for different, substantive complicity-avoidance doctrines—rules of substantive law that alter the content of a given cause of action in a way that thwarts private parties’ attempts to use civil litigation to achieve certain repugnant ends. Attend­ing to these doctrines not only reveals a deep connection between procedure and substance but also gives a fuller sense of the kinds of pri­vate wrongdoing that trigger complicity concerns. This section considers two substantive complicity-avoidance doctrines: the version of the state action doctrine applied in Shelley v. Kraemer and Seana Shiffrin’s revision­ist account of the unconscionability doctrine in contract law.

Shelley held that a State court’s enforcement of a racially restrictive covenant is “state action” that violates the Fourteenth Amendment’s Equal Protection Clause. 243 Shelley v. Kraemer, 334 U.S. 1, 14–20 (1948). While the Court relied on precedents holding that official judicial action is state action, 244 See id. at 14. it never clearly explained how private parties’ discriminatory motives can taint a judicial decision. The Court’s reasoning, however, indicates that the constitutional violation in Shelley must have amounted to state complicity in private discrimination. The Court never suggested that either of the States involved had itself engaged in any discriminatory conduct. Rather, the State courts had faci­litated private discrimination by giving effect to discriminatory private agreements. Their responsibility for the discrimination was therefore secondary rather than primary. This is reflected in the language of the Court’s opinion, which described the State courts’ enforcement of the restrictive covenants as the “[p]articipation of the State” in private discrimination. 245 Id. at 13. And while acknowledging that “the particular pattern of discrimination . . . was defined initially by the terms of a private agree­ment,” the Court noted that “[t]he judicial action” of enforcing that discrimination “bears the clear and unmistakable imprimatur of the State.” 246 Id. at 20. Such language describes the role of the accomplice, not the principal.

The private wrongdoing that the State courts had facilitated was a matter of the substance of the restrictive covenants, rather than anything the parties had done during the litigation to enforce them. The Court accordingly had to look to substantive law to develop a strategy for avoiding state complicity in that wrongdoing. Given that the racially restrictive covenants were enforceable under each State’s common law of property, the Court essentially ordered the States to revise their common law to no longer recognize such agreements, 247 See id. at 19, 22. thereby making the agreements solely a matter of “voluntary adherence.” 248 Id. at 13. By refusing to coercively enforce the racially restrictive covenants, the Court believed, the States could extricate themselves from that private discrimination. That is the logic of complicity avoidance.

This account by no means solves the perennial puzzle surrounding Shelley: how to cabin its holding to prevent constitutional law from subsuming nearly all of private law. The Supreme Court has itself balked at Shelley’s most sweeping implications, significantly reining in the state action doctrine in subsequent decisions. 249 See Davis, supra note 18, at 594 n.40 (explaining that Shelley was the closest the Court has come to treating “every private action as state action” but further noting that the Court “has never read that case for all it may be worth”). But the Court and commenta­tors alike have failed to articulate a coherent, principled basis for limiting Shelley’s logic. 250 See, e.g., Sunstein, Partial Constitution, supra note 47, at 56, 73 (noting that Shelley is “extremely controversial” and that the “puzzle of the case” lies in the fact that “judicial enforcement of voluntary agreements is not normally thought to violate the Constitution”); Laurence H. Tribe, American Constitutional Law 1697 (2d ed. 1988) (arguing that the Court’s modern state action decisions “appear peculiarly unpersuasive” in purporting to “articulate and apply an autonomous state action doctrine”); Mark Tushnet, Weak Courts, Strong Rights: Judicial Review and Social Welfare Rights in Comparative Constitutional Law 167 (2008) (explaining that the rule established in Shelley can be difficult to apply in other state action cases); cf. Charles L. Black, Jr., The Supreme Court, 1966 Term—Foreword: “State Action,” Equal Protection, and California’s Proposition 14, 81 Harv. L. Rev. 69, 95 (1967) (calling the state action doctrine “a conceptual disaster area”). In light of that failure, we could simply dismiss Shelley as a (necessary and justified) realist aberration, the high-water mark of the Court’s attempt to extend federal judicial oversight over as many discrim­inatory practices as possible. A less ad hoc response would be to accept the full breadth of Shelley’s articulation of the state action doctrine and to focus on determining when the state should refuse to lend its support to private wrongdoing. It may be better to candidly acknowledge the extent of the state’s potential involvement in private wrongdoing and to consi­der when the state has a duty to repudiate it than to occlude the moral costs of our commitment to private ordering by simply pretending that the state isn’t implicated. 251 Cf. Sunstein, Partial Constitution, supra note 47, at 160 (“Shelley v. Kraemer . . . was a remarkably easy state action case. . . . The case was difficult only because it is unclear whether the Constitution forbids the state’s apparently neutral use of its courts to enforce contracts, including racially restrictive property agreements.”).

Read as a complicity-avoidance decision, Shelley identifies a problem that, far from being limited to constitutional law’s state action doctrine, recurs throughout the law. Consider the unconscionability doctrine in contract law, as reconceptualized by Shiffrin. 252 See generally Shiffrin, Paternalism, supra note 25. For a recent critique of Shiffrin’s account, see Nicolas Cornell, A Complainant-Oriented Approach to Unconscionability and Contract Law, 164 U. Pa. L. Rev. 1131, 1153–57 (2016). The unconscionability doctrine allows courts to refuse to enforce an otherwise valid contract the terms of which were manifestly unfair or immoral at the time the agree­ment was formed. 253 See U.C.C. § 2-302(1) (Am. Law Inst. & Unif. Law Comm’n 2003); Restatement (Second) of Contracts § 208 (Am. Law Inst. 1981). Both critics 254 See, e.g., Charles Fried, Contract as Promise 104 (1981); Alan Schwartz, A Reexamination of Nonsubstantive Unconscionability, 63 Va. L. Rev. 1053, 1071–76 (1977). and defenders 255 See, e.g., Duncan Kennedy, Distributive and Paternalist Motives in Contract and Tort Law, with Special Reference to Compulsory Terms and Unequal Bargaining Power, 41 Md. L. Rev. 563, 631–38 (1982); Anthony T. Kronman, Paternalism and the Law of Contracts, 92 Yale L.J. 763, 764 (1983). of the doctrine have viewed it as paternalistic, since it relieves one of the parties to a con­tract of her freely assumed obligations when they prove to have been unwise. Against this consensus, Shiffrin contends that the unconscionability doc­trine need not be paternalistic, depending on the state’s motive in deploy­ing it. Insofar as the state uses the doctrine simply to protect the weaker party to a contract from the bad consequences of her decision, then it is indeed paternalistic. But that is not the state’s only possible purpose. Instead, the state’s “motive may reasonably be a self-regarding concern not to facilitate or assist harmful, exploitative, or immoral action.” 256 Shiffrin, Paternalism, supra note 25, at 224. This “self-regarding concern” arises because “the legal institu­tion of contract requires, through their role in enforcement, the participation and cooper­ation of parties outside the agreement—that is, it requires the cooperation of the community, as it is embodied in the state.” 257 Id. at 221. So rather than bespeaking an other-regarding concern for the weaker party’s well-being, the “state’s refusal to enforce an unconscionable contract could reflect an unwillingness to lend its support and its force to assist an exploitative contract because it is an unworthy endeavor to support.” 258 Id. at 223–30, 249–50. One sees this alternative motivation at work in many of the decisions applying the doctrine, in which “a dominant concern of judges is self-regarding: it is to avoid facilitating the actions of an exploiter rather than to act to protect the disadvantaged party.” 259 Id. at 229 (citing illustrative cases). That, of course, is precisely the same concern as in Shelley. 260 Shiffrin briefly makes the connection to Shelley. See id. at 233 n.34. Shiffrin thus reconceptualizes unconscionability as a complicity-avoidance doctrine. 261 One can give a similar account of other contract law doctrines that allow courts to refuse to enforce private agreements on grounds of “public policy.” Beyond contract law, another potential example of a substantive complicity-avoidance doctrine is the equitable “clean hands” defense, which has traditionally been justified as a way for courts to avoid abetting or countenancing private wrongdoing and to thereby preserve their “integrity.” See Ori J. Herstein, A Normative Theory of the Clean Hands Defense, 17 Legal Theory 171, 176–78 (2011). But see id. at 188–89 (doubting whether the kinds of wrongdoing that typically trigger the clean hands defense rise to the level of implicating the political community).

While substantive, both Shelley and unconscionability respond to the same fundamental problem as civil procedure’s delegation-policing doctrines. Whether by abusing delegated procedural powers or by seeking public support for their immoral projects, private parties can render the state complicit in their wrongdoing. Shelley and unconsciona­bility, just like civil procedure’s delegation-policing doctrines, enable the state to avoid the worst forms of complicity.

C. Complicity Avoidance, Not Due Process

If the logic of complicity avoidance underlies important doctrines in civil procedure (among other legal domains), one might wonder why it has escaped scholarly attention. The neglect may well owe to the hege­mony of another normative logic in civil procedure doctrine and scholar­ship: procedural due process. In answering almost any procedural question, courts and scholars alike reflexively invoke due process—if not the concept itself, then at least its constituent values, such as fairness and accuracy. Many readers thus might doubt the distinctiveness of complicity avoidance and instead try to assimilate it to due process. There is consid­erable support for such a move, as both courts and commentators have conceptualized delegation problems in other areas of the law in due process terms. 262 See, e.g., Carter v. Carter Coal Co. 298 U.S. 238, 326, 330–32 (1936) (analyzing the validity of the price-setting provisions in the Bituminous Coal Conservation Act of 1935 under the Fifth Amendment’s Due Process Clause); Davis, supra note 18, at 615–21 (identifying due process limits on private enforcement of regulatory laws); Lawrence, supra note 42, at 672–95 (considering “the due process approach for handling private delegation”); Rubenfeld, supra note 39, at 11–19 (discussing the applicability of the Due Process Clause to Title IX investigations and adjudications conducted by private schools). Yet, while civil procedure’s delegations of state power can indeed raise significant due process concerns, those concerns hardly exhaust the delegations’ normative significance. Even after accounting for the threat to due process values, there remains the problem of state complicity in abusive exercises of delegated power—a problem that civil procedure’s delegation-policing doctrines are well tailored to address.

Procedural due process has two main features, neither of which quite fits the structure of civil procedure’s delegation problem. First, procedural due process reflects an ex ante concern with promoting accuracy in governmental decisionmaking. The Supreme Court has construed the Due Process Clause to require the government to follow certain procedures before depriving people of their property (or, less relevant in the civil context, their lives or liberty). 263 See, e.g., Fuentes v. Shevin, 407 U.S. 67, 80 (1972). Thus, the question in the canonical procedural due process cases of Mathews v. Eldridge, 424 U.S. 319 (1976), and Goldberg v. Kelly, 397 U.S. 254 (1970), was whether recipients of certain federal benefits were entitled to a hearing before the government terminated their benefits. According to the Court, the purpose of this requirement is “to minimize substantively unfair or mistaken deprivations of property” by the government. 264 Fuentes, 407 U.S. at 81. Procedural due process thus seeks to prevent erroneous deprivations of property by prescribing procedures that enhance the accuracy of govern­mental decisionmaking. 265 See, e.g., Lawrence A. Alexander, Cutting the Gordian Knot: State Action and Self-Help Repossession, 2 Hastings Const. L.Q. 893, 917–18 (1975) (“The basic purpose of the procedural due process requirements associated with the right to a hearing on adjudicative facts . . . is to minimize error in the finding of adjudicative facts.”). Some scholars have attempted to ground due process in values other than accuracy. See, e.g., Jerry L. Mashaw, Due Process in the Administrative State 158–253 (1985) (developing a dignitarian theory of due process); Lawrence B. Solum, Procedural Justice, 78 S. Cal. L. Rev. 181, 273–84 (2004) (arguing that participatory procedures enhance the legitimacy of governmental decisions). These alternative conceptions of due process, however, can still be distinguished from complicity avoidance on the second ground considered in this section.

Civil procedure’s delegation-policing doctrines, by contrast, reflect an ex post concern that goes beyond accurate decisionmaking. Even when a civil justice system institutes procedures that reduce the risk of erroneous outcomes to a tolerable level, those procedures will not forestall every abuse of delegated state power. Nor will every kind of abuse compromise the accuracy of the decisionmaking process. Yet civil procedure still employs delegation-policing doctrines, which seek to address abusive exercises of delegated state power as they are occurring or after they have occurred, and irrespective of their effect on the accuracy of the ultimate decision. Traditional notions of due process can’t readily account for that function; complicity avoidance can.

Once again, civil Batson exemplifies the divergence between civil pro­cedure’s delegation-policing doctrines and procedural due process. The Court in Edmonson admittedly adverted to due process concerns in justifying a ban on racially discriminatory peremptory challenges in civil trials, noting how “[r]ace discrimination within the courtroom raises serious questions as to the fairness of the proceedings conducted there.” 266 Edmonson v. Leesville Concrete Co., 500 U.S. 614, 628 (1991). It is also true that the traditional justifications for peremptory challenges sound in due process. 267 See id. at 633 (O’Connor, J., dissenting) (“By allowing the litigant to strike jurors for even the most subtle of discerned biases, the peremptory challenge fosters both the perception and reality of an impartial jury.”). But Edmonson’s holding is both underinclusive and overinclusive with respect to due process values. Rather than specify ex ante various accuracy-undermining bases for making peremptory challenges, the Court subjected peremptory challenges to ex post review solely for discrimination. And rather than permit race-based peremptory challenges that may be grounded in empirically justified generalizations, the Court proscribed all race-based peremptory challenges categori­cally. 268 Id. at 630–31. These aspects of Edmonson’s holding belie the suggestion that the Court was trying to enhance the overall accuracy of civil jury trials. They instead suggest that the Court was trying to avoid state complicity in pri­vate discrimination as such.

Second, procedural due process is concerned with direct state action, whereas civil procedure’s delegation-policing doctrines are concerned with indirect state action. Due process constrains what the state may do to private parties—the delegation-policing doctrines, what it may do for private parties. Consider the prejudgment attachment and replevin cases. Those cases were formally adjudicated under the Due Process Clause, and the Court repeatedly emphasized the importance of predeprivation hearings for promoting accurate decisionmaking. 269 See, e.g., Connecticut v. Doehr, 501 U.S. 1, 13 (1991); Fuentes, 407 U.S. at 80–82. But the Court also acknowledged that while in those cases “the inquiry is similar” to the traditional due process analysis, “the focus is different.” 270 Doehr, 501 U.S. at 10. Specifically, “[p]rejudgment remedy statutes ordinarily apply to disputes between private parties rather than between an individual and the government,” and as such “are designed to enable one of the parties to ‘make use of state procedures with the overt, significant assistance of state officials.’” 271 Id. at 10–11 (quoting Tulsa Prof’l Collection Servs., Inc. v. Pope, 485 U.S. 478, 486 (1988)). The state’s role, in other words, is limited to facilitating, rather than directly regulating, private conduct. The Court thus had to gerrymander the three-pronged due process balancing test articulated in Mathews v. Eldridge, 272 424 U.S. 319, 335 (1976). according “principal attention” to “the interest of the party seeking the prejudgment remedy” rather than to any interest of the government. 273 Doehr, 501 U.S. at 11. More than just a minor doctrinal tweak, this modification fundamentally reconfigured the role of the state in the due process analysis. The concept of complicity avoidance makes that reconfiguration seem less ad hoc.

For there to be state complicity, of course, there must be some under­lying private wrong for the state to be complicit in. And one can indeed conceptualize some of the wrongs addressed by civil procedure’s delegation-policing doctrines in due process terms, as some involve, or contribute to, an arbitrary deprivation of property. But not all of them. From civil Batson to Rule 11 to the abuse-of-process tort, the problem is not that someone was or might be arbitrarily deprived of her property—some of the doctrines can be triggered without any deprivation at all—but that a private party has compromised certain public values in exer­cising state power. In any event, even when the underlying wrong does implicate due process, the state’s role remains only secondary rather than primary. The state does not act of its own initiative but rather lends its coercive apparatus to private parties so they can achieve their own ends. Complicity avoidance provides a distinct normative framework for thinking about the state’s obligations when it has delegated its coercive power to private parties rather than exercising that power itself.

D. Defending Complicity Avoidance—and Delegation

The complicity-avoidance logic employed in cases such as Edmonson is not just a doctrinal construct. As this section argues, that logic reso­nates with fundamental principles of liberal political theory. The aim, to be clear, is not to provide a definitive defense of complicity avoidance but to show it to be a normatively attractive ideal in light of our legal system’s other commitments. At the same time, complicity avoidance comes at a cost: In attempting to avoid complicity in private wrongdoing, the state forgoes many of the benefits associated with delegating state power to private parties in the first place. The upshot, this section suggests, is that sometimes there are good reasons for the state—and thus the entire politi­cal community—to tolerate a degree of complicity in private wrongdoing.

1. The Political Theory of Complicity Avoidance. — Civil procedure’s delegations of state power run up against one of the most basic tenets of liberal political theory: that the defining feature of political power is its coerciveness and that the central task of political theory is legitimating and limiting that coercion. Contemporary liberalism proceeds from the premise, most prominently associated with Max Weber, that the modern state claims a monopoly on the legitimate use of physical force. 274 See Max Weber, The Profession and Vocation of Politics (1919), reprinted in Weber: Political Writings 310 (Peter Lassman & Ronald Speirs eds., 1994). This premise gives rise to the need to specify the conditions that legitimate the state’s use of coercion, which, in turn, involves identifying limits on the state’s coercive power. 275 See, e.g., Ronald Dworkin, Law’s Empire 191 (1986) (describing “the classical problem of the legitimacy of coercive power”). These imperatives explain public law’s enduring focus on constraining state action. 276 See Amitai Etzioni, Reining in Private Agents, 101 Minn. L. Rev. Headnotes 279, 279 (2016), [] (“The Constitution and much of statutory law seek to protect individual rights from intrusions by the government. It is the government’s coercive power that must be constrained.”); Daryl J. Levinson, The Supreme Court 2015 Term—Foreword: Looking for Power in Public Law, 130 Harv. L. Rev. 31, 46 & n.79 (2016) (arguing that a primary ambition of constitutionalism is to constrain state power). They also account for the more recent anxiety about the “privatization” of various state functions 277 See supra notes 1–2, 140 and accompanying text. : If state power is necessarily coercive, and if that power must be subject to certain limits to be legitimate, then delegating state power to private par­ties threatens to untether state coercion from its legitimating constraints. All of this is reason enough to carefully attend to civil procedure’s delegations of state power and its delegation-policing doctrines.

A concern with coercion alone, however, doesn’t fully explain courts’ additional concern with complicity in abusive exercises of delegated state power. To appreciate that further worry, one must understand the special nature of political power in a liberal democracy. The political philosopher John Rawls, synthesizing the liberal social-contract tradition, famously argued that democratic political power is distinctive not just because of its coerciveness but also because of its authorship. “[W]hile political power is always coercive power,” Rawls explained, “in a constitutional regime it is the power of the public, that is, the power of free and equal citizens as a collective body.” 278 John Rawls, Political Liberalism 68 (1996) [hereinafter Rawls, Political Liberalism]. In a democracy, in other words, political power is ulti­mately “citizens’ power” 279 John Rawls, Justice as Fairness: A Restatement 40–41 (Erin Kelly ed., 2001). —“a power in which all citizens have an equal share.” 280 Id. at 183–84. This conception of political power differs fundamentally from that of genuine libertarianism (as distinct from “classical” versions of liberalism), which reduces political power to a “network of private contracts” and accordingly denies the existence of any separate “public” realm. Samuel Freeman, Illiberal Libertarians: Why Libertarianism Is Not a Liberal View, 30 Phil. & Pub. Aff. 105, 107 (2001). For liberals, by contrast, the establishment of political institutions constitutes a distinct public domain subject to a distinct set of normative principles. Cf. Samuel Scheffler, Membership and Political Obligation, 25 J. Pol. Phil. 3, 16 (2017) (“[T]he very idea of a ‘public’ is a creation of political society; there is no public in the state of nature.”). It is thus doubly public: As coercive power, it’s vested with the state, per the monopoly of physical force; and as democratic power, it’s col­lectively authorized by the political community, per the social contract. 281 But see James Q. Whitman, Between Self-Defense and Vengeance/Between Social Contract and Monopoly of Violence, 39 Tulsa L. Rev. 901, 902 (2004) (presenting the social contract and monopoly of violence as two opposing theories of the state).

This means that the political community has a lot riding, morally speaking, on how the state exercises its coercive power. As Arthur Ripstein has explained, because political power in a democracy is authorized by the political community as a whole, the state speaks in the name of the political community whenever it exercises power through any of its institutions. 282 See Arthur Ripstein, Private Order and Public Justice: Kant and Rawls, 92 Va. L. Rev. 1391, 1435 (2006) [hereinafter Ripstein, Private Order]. That includes when courts resolve private disputes through civil litigation, which is among those “procedures and institutions [that] can decide to act on behalf of everyone.” 283 Id. at 1428; see also Ripstein, Private Wrongs, supra note 51, at 289 (“A court resolving a private dispute purports to speak from . . . a public standpoint . . . .”); Shiffrin, Remedial Clauses, supra note 6, at 420 (“[T]he judiciary serves as the community’s voice, levying its response to an abrogation of a publicly recognized responsibility.”). The political community thus bears ultimate responsibility for how the state exercises its coercive power, including how its courts adjudicate private disputes. 284 See Ripstein, Private Order, supra note 282, at 1435. And when the state abuses its power, the political community becomes complicit in that wrongdoing. 285 For more on the idea that the citizens of a democracy collectively authorize, and thus bear responsibility for, many of the acts of their government, see generally Eric Beerbohm, In Our Name: The Ethics of Democracy (2012); John M. Parrish, Collective Responsibility and the State, 1 Int’l Theory 119 (2009); Anna Stilz, Collective Responsibility and the State, 19 J. Pol. Phil. 190 (2011). But see Robert Jubb, Participation in and Responsibility for State Injustices, 40 Soc. Theory & Prac. 51, 52–53 (2014) (criticizing this idea).

Yet the liberal account of political power has much more sweeping implications for the civil justice system than liberal political theorists have appreciated. They tend to equate exercises of state power in civil litiga­tion with judicial action, and thus assume that the political community’s complicity extends only as far as the decisions of courts. But as this Article has shown, the state doesn’t maintain exclusive control over its coercive power in civil litigation. To be sure, as liberal political theorists have rightly emphasized, the main point of a civil justice system is to provide authoritative, public resolutions of private disputes. 286 See Ripstein, Private Order, supra note 282, at 1415–29. Along the way, though, the state permits private parties to exercise various procedural powers. 287 See supra section I.B. As coercive powers of a democratic state, those powers are still authorized by the members of the political community and exercised in their name. The political community thus remains collectively respon­sible for how private parties exercise the delegated state powers during civil litigation. Given the nature of democratic political power and the division of labor between private parties and courts in civil litigation, the complicity that concerns courts in cases such as Edmonson turns out to be not just the complicity of the state but the complicity of us all. 288 This Article’s complicity-avoidance theory of civil procedure’s delegation-policing doctrines shares certain affinities with Malcolm Thorburn’s public law theory of criminal law’s justification defenses. See Malcolm Thorburn, Criminal Law as Public Law, in Philosophical Foundations of Criminal Law 21 (R.A. Duff & Stuart P. Green eds., 2011); Malcolm Thorburn, Justifications, Powers, and Authority, 117 Yale L.J. 1070, 1118 (2008) [hereinafter Thorburn, Justifications]. But whereas Thorburn focuses on emergency situations in which private parties must temporarily “stand in” for state officials, see Thorburn, Justifications, supra, at 1076, 1107, 1118, 1125–26, this Article contends that civil procedure delegates coercive state power to private parties on a continuing, rather than emergency, basis, and that private parties don’t “stand in” for the state but rather are permitted to use delegated state power primarily for their own purposes.

2. Balancing Complicity Avoidance and Delegation. — If the political community can indeed become complicit in private exercises of delegated state power during civil litigation, then perhaps we should just eliminate the risk of complicity by eliminating the delegations. Rather than delegate the state’s coercive prosecutorial, investigatory, and adjudi­catory powers to potentially faithless private parties, the safer course might be to vest those powers exclusively with more accountable governmental officials. And yet, while civil procedure’s delegations entail significant moral costs for the state—much more significant than civil procedure scholars have appreciated—they also promise significant benefits. Those benefits counsel against eliminating civil procedure’s delegations and in favor of taking the very approach civil procedure in fact takes: delegating many of the state’s procedural powers to private parties but policing those delegations for abuse.

Many of the benefits of civil procedure’s delegations have been well canvassed in the literature and are present across the civil docket, both in private law and public law cases. In private law cases, delegating the civil prosecutorial power to private parties empowers the victims of private wrongs to directly vindicate their rights 289 This is one of the main insights of civil recourse theory. See Goldberg & Zipursky, Torts as Wrongs, supra note 50, at 981–82 (arguing that civil recourse “enables individuals to assert claims as a matter of right without first obtaining the permission or blessing of government officials”); cf. Nagareda, supra note 18, at 191 (“In general, one of the advantages of civil litigation lies in the ability of a private person to lodge a complaint in a manner independent from the government.”). and reposes prosecutorial discretion with those individuals who experience the costs and benefits of civil litigation most acutely. 290 See Shiffrin, Remedial Clauses, supra note 6, at 433 (“[W]ith the duties of ‘private’ law, because private parties bear the brunt of their burdens as well as the effects of their violation, we generally assign private parties the power to pursue their vindication, despite the underlying, concomitant public interest.”). In cases with a public law dimension, schol­ars have recognized a number of additional benefits associated with the private enforcement of federal statutes, mostly stemming from various practical advantages that private parties enjoy over administrative agencies, including less vulnerability to capture, stronger enforcement incentives, more resources, and greater access to information. 291 For a good overview of these benefits, see Clopton, Redundant, supra note 14, at 308–11; see also, e.g., Pamela H. Bucy, Private Justice, 76 S. Cal. L. Rev. 1, 54–60 (2002) (considering the investigative and informational advantages of private parties); Stephen B. Burbank, Sean Farhang & Herbert M. Kritzer, Private Enforcement, 17 Lewis & Clark L. Rev. 637, 662–67 (2013) (evaluating commonly made arguments regarding the benefits of private enforcement, including increasing prosecutorial resources, shifting costs from public to private sources, and “facilitat[ing] participatory and democratic governance”); Margaret H. Lemos & Max Minzner, For-Profit Public Enforcement, 127 Harv. L. Rev. 853, 912–13 (2014) (noting that “[f]inancial incentives blur the line between public and private enforcement”); Matthew C. Stephenson, Public Regulation of Private Enforcement: The Case for Expanding the Role of Administrative Agencies, 91 Va. L. Rev. 93, 106–20 (2005) (discussing the potential advantages and disadvantages of private enforcement). Scholars have ascribed similar benefits to party-driven discovery. See, e.g., Carrington, supra note 96, at 55 (suggesting that discovery has helped to enable private enforcement and reduce demand for government or bureaucratic intervention); Miller, Simplified Pleading, supra note 77, at 356 (arguing that limits on discovery should be calibrated so as to preserve “the civil enforcement role” performed by private litigation). Beyond practical considerations, private enforcement responds to Americans’ traditional ideological distrust of governmental power. See generally Robert Kagan, Adversarial Legalism: The American Way of Law (2001) (connecting the United States’ preference for private enforcement to Americans’ suspicion of the state).

Delegating state power has another set of interrelated benefits that have more to do with political morality than efficient policy implementa­tion. Scholars have explored these benefits in the context of settlement and various forms of alternative dispute resolution (ADR), but they seem to apply to the private exercise of other state procedural powers as well. Most directly, delegating state procedural powers to private parties pro­motes individual autonomy by giving individuals some say over the resolution of their disputes. 292 See Carrie Menkel-Meadow, Whose Dispute Is It Anyway?: A Philosophical and Democratic Defense of Settlement (in Some Cases), 83 Geo. L.J. 2663, 2692 (1995) (“Settlements that are in fact consensual represent the goals of democratic and party-initiated legal regimes by allowing the parties themselves to choose processes and outcomes for dispute resolution.”); see also Edward Brunet, The Core Values of Arbitration, in Edward Brunet et al., Arbitration Law in America: A Critical Assessment 3 (2006) (making a similar autonomy-based argument for arbitration). Because of civil procedure’s delegations, private parties can (within wide limits) delineate the contours of their dispute in terms of claims and parties, decide how much and what kinds of information they will exchange with each other during discovery, and even set the terms of any resolution itself through settlement. This autonomy, in turn, promises two further benefits. First, rather than stoke atomism, allowing private parties to exercise the state’s procedural powers might actually foster communal solidarity, on the theory that people develop a closer attachment to the law by enforcing it them­selves. 293 See Menkel-Meadow, supra note 292, at 2688–89 (arguing that settlement can allow for greater participation, and thus a closer connection to the law, than litigation); cf. Jody Freeman, Extending Public Law Norms Through Privatization, 116 Harv. L. Rev. 1285, 1285 (2003) (arguing that “privatization can be a means of ‘publicization,’ through which private actors increasingly commit themselves to traditionally public goals”). Second, delegation is also a way for the state to accommodate pluralism. Some members of a liberal political community will inevitably espouse values different from those embodied in the law and other political institutions. 294 This results from what Rawls called the “fact of reasonable pluralism.” Rawls, Political Liberalism, supra note 278, at 36. By delegating its procedural powers, the state gives private parties space to express those values in the course of resolving their disputes. 295 See Menkel-Meadow, supra note 292, at 2672–78 (arguing that settlement allows disputes to be decided according to values other than those embodied in the law); see also Robert M. Ackerman, Disputing Together: Conflict Resolution and the Search for Community, 18 Ohio St. J. on Disp. Resol. 27, 67–69 (2002) (identifying various communitarian benefits of arbitration). This is one of the main justifications given for religious arbitration and, more generally, arbitration of disputes between the members of a close-knit community. See, e.g., Michael A. Helfand, Arbitration’s Counter-Narrative: The Religious Arbitration Paradigm, 124 Yale. L.J. 2994, 3010–22 (2015) (discussing how arbitration can “advance a shared set of objectives and values”).

In light of delegation’s benefits, the state might be drawn toward the opposite extreme. Rather than avoid complicity in private wrongdoing by eliminating civil procedure’s delegations and exercising all coercive procedural powers itself, the state might try to get out of the dispute-resolution business altogether, leaving it exclusively to private ordering. 296 Cf. Resnik, Contract, supra note 6, at 644 (considering a “model of disengage­ment” whereby judges would extricate themselves from settlement negotiations). This strategy, however, runs up against another core liberal commitment: the state’s obligation to provide a forum for the authoritative resolution of private disputes. 297 See supra note 286 and accompanying text; see also, e.g., Radin, Threat, supra note 46, at 295–97 (describing liberalism’s conception of the state’s role in resolving private disputes). Indeed, liberalism has traditionally viewed the adju­dication of private disputes as a way to augment the state’s authority and enhance its legitimacy. 298 See, e.g., Stephen Holmes, The Anatomy of Antiliberalism 201–10 (1993); cf. Judith Resnik & Dennis E. Curtis, Representing Justice: Invention, Controversy, and Rights in City-States and Democratic Courtrooms 338 (2011) (showing how throughout history courthouses and other sites of public dispute resolution have symbolized state authority). Liberalism thus creates a dilemma for the state: If the state attempts to avoid complicity in private wrongdoing by exercising all coercive power itself, it forgoes the considerable benefits of delegation; but if it attempts to avoid complicity by completely out­sourcing dispute resolution, it shirks its role as final arbiter.

This Article does not attempt to resolve that dilemma. Neither the delegation framework developed in Parts I and II nor the complicity-avoidance theory developed in this Part can tell us exactly how much power a civil justice system should delegate to private parties. Given all the competing imperatives, however, it does seem that our current civil justice system strikes a reasonable balance. Civil procedure’s delegations realize a number of practical and moral benefits for individual litigants and the political community at large, while its delegation-policing doctrines help the political community to avoid becoming complicit in the worst forms of abuse of those delegations. One can, of course, readily envision perfectly legitimate civil justice systems that strike a different balance—either delegating less state power to private parties to avoid even more complicity in private wrongdoing or delegating even more state power to better promote autonomy and accommodate pluralism. The important point is to recognize the tradeoff between complicity avoidance and other important values served by the delegation of state power. Our civil justice system accepts a significant degree of public complicity in private wrongdoing for the sake of those other values, rather than seeking to minimize any threat to the political community’s moral integrity by maintaining exclusive state control over coercive state power.

IV. Toward a Delegation-Policing Approach to Arbitration

The analytical framework developed in the previous Parts can shed light on several procedural debates, both within ordinary civil litigation and beyond it. Within ordinary civil litigation, this Article’s delegation framework simultaneously highlights some of the benefits and drawbacks of current procedural practice. For example, although judges enjoy substantial discretion to police discovery abuse, there is a widespread impres­sion that they rarely exercise it, with the result that discovery often ends up devolving into a “free-for-all.” 299 Pollis, supra note 91, at 2099. The delegation framework sug­gests that judges can respond to such problems (insofar as they’re real problems) 300 Cf. Danya Shocair Reda, The Cost-and-Delay Narrative in Civil Justice Reform: Its Fallacies and Functions, 90 Or. L. Rev. 1085, 1088–90 (2012) (arguing that the costs of discovery are exaggerated); Spencer, supra note 157, at 1730 (same). by more aggressively policing discovery for abuse ex post, in contrast to the recent amendments to the discovery rules, which focus on limiting the scope of discovery ex ante. 301 See supra note 91 and accompanying text.

The delegation framework also speaks to delegations of the state’s dispute-resolution powers outside ordinary civil litigation. Such delega­tions include so-called “corporate settlement mills,” a practice in which corporations that have been accused of wrongdoing essentially adjudi­cate victims’ claims, sometimes in ways that preclude the victims from seeking relief in court. 302 Dana A. Remus & Adam S. Zimmerman, The Corporate Settlement Mill, 101 Va. L. Rev. 129, 136–41 (2015) (arguing that corporate settlement mills “privatize dispute resolution nearly completely, and are subject to far weaker (if any) forms of public regulation and oversight”). This Article’s framework confirms the many concerns that scholars have voiced about this practice and suggests the outlines of a general approach to enhancing its legitimacy.

But however important these and other procedural issues may be, the most pressing one remains arbitration. This Part accordingly revisits the delegation critique of arbitration in light of what the previous Parts have revealed about civil procedure’s own delegations of state power. As section IV.A argues, proponents of the delegation critique rightly worry that arbitration delegates significant coercive power to private parties and that, under current Supreme Court precedent, private parties can abuse that power with virtual impunity. In the end, though, the delegations of state power in arbitration are conceptually continuous with those in ordinary civil litigation; the former may lend themselves to more egre­gious forms of abuse, but both sets of delegations fundamentally raise the same kinds of concerns. This continuity suggests an alternative approach to arbitration’s delegation problem: adapting civil procedure’s delega­tion-policing doctrines for arbitration’s delegations.

Section IV.B begins to take up that task. This Article’s approach focuses less on the enforceability of arbitration clauses and more on the reviewability of arbitral proceedings and awards. Whereas holding broad classes of arbitration clauses unenforceable effectively restricts private parties’ access to arbitration’s delegations before any abuse has occurred, reviewing arbitral proceedings and awards allows courts to police those delegations for abuse after the fact—the primary strategy civil procedure employs with respect to the delegations in ordinary civil litigation. There may indeed be compelling reasons to severely restrict access to arbitration’s delegations ex ante. But short of that, the lesson of civil procedure’s delegations is that many of the concerns about delegated state power in arbitration can be addressed by policing arbitration’s delegations for abuse ex post.

A. Delegating Procedure Redux

1. The Delegation Critique Defended. — Proponents of the delegation critique tend to portray arbitration as a radical departure from ordinary civil litigation. Whereas public institutions and officials discharge the state’s dispute-resolution functions in ordinary civil litigation, most scholars suppose, arbitration “privatizes” or “outsources” many of those functions. 303 See supra notes 6–9, 31–33 and accompanying text. We can now see that this contrast is overdrawn. 304 And not just because of practices such as contract procedure and court-annexed ADR. See supra notes 10, 26. Ordinary civil litigation likewise delegates many of the state’s dispute-resolution functions to private parties, thereby vesting them with significant coercive power.

This is not to say that scholars have been wrong to worry about the delegations of state power involved in arbitration. On the contrary, those delegations raise the same basic concerns as the delegations in ordinary civil litigation. To be sure, what this Article calls the “delegation critique” actually comprises a number of related, but distinct, criticisms, including the following: Binding dispute resolution is an inherently “public function” that should not be delegated to private parties except subject to strict limits, 305 See, e.g., Reuben, Constitutional Gravity, supra note 9, at 997–99; cf. Davis & Hershkoff, supra note 10, at 533 (raising similar concerns about contract procedure). Resnik’s concerns about “privatization” and “outsourcing” can also be understood in these terms. See supra note 6 and accompanying text. private parties should not be able to use procedure to insulate themselves from liability for violating substantive rights, 306 See, e.g., Radin, Fine Print, supra note 8, at 33–34; Glover, supra note 8, at 3075–76; David S. Schwartz, Claim-Suppressing Arbitration: The New Rules, 87 Ind. L.J. 239, 240–42 (2012) [hereinafter Schwartz, Claim-Suppressing Arbitration]; Schwartz, Small Print, supra note 8, at 110. and private parties should generally be prohibited from making law. 307 See, e.g., Horton, Delegation, supra note 7, at 480; David Horton, Mass Arbitration and Democratic Legitimacy, 85 U. Colo. L. Rev. 459, 501–02 (2014). Scholars have further argued that adjudication not only resolves private disputes but also realizes a number of public goods that arbitration forgoes—most notably, the production of legal precedent to guide the resolution of future cases 308 See, e.g., Richard M. Alderman, Consumer Arbitration: The Destruction of the Common Law, 2 J. Am. Arb. 1, 11–12 (2003); Myriam Gilles, The Day Doctrine Died: Private Arbitration and the End of Law, 2016 U. Ill. L. Rev. 371, 409–13; Charles L. Knapp, Taking Contracts Private: The Quiet Revolution in Contract Law, 71 Fordham L. Rev. 761, 785 (2002); Rex R. Perschbacher & Debra Lyn Bassett, The End of Law, 84 B.U. L. Rev. 1, 30 (2004). This criticism echoes Owen Fiss’s classic critique of settlement. See Owen M. Fiss, Against Settlement, 93 Yale L.J. 1073, 1085 (1984) (arguing that settlement stymies the development of the law). and public transparency about violations of substantive law. 309 See, e.g., Resnik, Diffusing Disputes, supra note 6, at 2894–900; Sternlight, Mandatory Arbitration, supra note 27, at 1664–65; cf. Davis & Hershkoff, supra note 10, at 513–14 (raising similar concerns about contract procedure). And whereas civil litigation proceeds (largely) in open court, arbitration is almost completely shielded from public scrutiny. 310 See, e.g., Resnik, Access, supra note 6, at 133–42; Judith Resnik, Courts: In and out of Sight, Site, and Cite: The Norman Shachoy Lecture, 53 Vill. L. Rev. 771, 799–809 (2008); Resnik, Uncovering, supra note 6, at 549–51; Judith Resnik, Whither and Whether Adjudication?, 86 B.U. L. Rev. 1101, 1137 (2006).

While it is important to recognize the differences among these various criticisms of arbitration, each ultimately presumes the kind of delegation identified in Part I. Each problem, that is, arises only because arbitration allows private parties to tap the coercive power of the state without any meaningful judicial oversight. Thus, private parties perform the inherently “public function” of dispute resolution only insofar as their decisions are binding—more or less automatically backed by the coercive power of the state. 311 See Reuben, Constitutional Gravity, supra note 9, at 997 (arguing that, in contrast to nonbinding dispute resolution, “[t]he binding resolution of disputes is, of course, a traditionally exclusive public function”). Private parties can use procedure to evade substantive legal obligations only insofar as courts will more or less automatically enforce their procedural choices. Privately made rules constitute “law” only insofar as they’re coercively enforced without any real judicial review. 312 See Horton, Delegation, supra note 7, at 441, 444 (arguing that the FAA, as construed by the Supreme Court, “allows private parties to engage in lawmaking” by “mak[ing] arbitration clauses enforceable with few restrictions” and “allow[ing] drafters to cut judges out of the loop”). Arbitration forgoes the public goods of precedent and transparency only insofar as private parties can authoritatively resolve their disputes unsupervised. And private parties can shield dispute reso­lution from public scrutiny only insofar as their authoritative decisions are insulated from judicial review, which would be subject to public scrutiny. None of these problems would be so acute if a party to an arbitration agreement could, notwithstanding the agreement, either litigate her dispute in court or at least obtain meaningful judicial review of any arbitral award; 313 Before the FAA, the “revocability doctrine” allowed a party to an arbitration agreement to revoke the agreement and litigate her dispute in court at any time before an arbitral award was rendered. See Ian R. Macneil, American Arbitration Law: Reformation, Nationalization, Internalization 20 (1992). each problem stems from the fact that the state coercively enforces arbitration without adequate scrutiny. This is the sense in which arbitration can be understood to delegate state power. 314 Cf. Davis & Hershkoff, supra note 10, at 535 (linking the concern that contract procedure “privatiz[es],” “outsource[es],” or “delegate[es]” the state’s dispute-resolution powers with the concern that it forgoes various “public goods” associated with litigation).

And the current arbitration regime does indeed pervasively delegate state power in precisely that sense. As others have recognized, the FAA, which governs the relationship between courts and arbitration, contains multiple delegations on its face. 315 See, e.g., Brunet, Arbitration and Constitutional Rights, supra note 9, at 109; Horton, Delegation, supra note 7, at 480; Reuben, Constitutional Gravity, supra note 9, at 1006; Sternlight, Rethinking, supra note 9, at 42. On the front end, the Act requires federal district courts, upon motion, to compel arbitration and to stay any related judicial proceedings in any dispute subject to a valid arbitration clause. 316 9 U.S.C. §§ 3–4 (2012). As for the arbitral proceedings themselves, the FAA authorizes the parties to choose a method for appointing an arbitrator and, by exten­sion, the procedures she will follow. 317 Id. § 5. If the arbitration clause fails to specify a method or if the parties fail to follow the specified method, then the court may appoint an arbitrator. Id. Whomever they appoint, in turn, is empowered to subpoena nonparty witnesses and documentary evidence. 318 Id. § 7. Courts are divided over whether this provision permits arbitrators to compel prehearing “discovery” from nonparties or only the production of documents and the appearance of witnesses at an actual arbitral “hearing.” See ­Randy J. Sutton, Annotation, Discovery in Federal Arbitration Proceedings, 45 A.L.R. Fed. 2d 51, §§ 13–16 (2010) (collecting cases). Although a subpoena issues “in the name of the arbitrator” and under her signature, the parties can petition a federal district court for an order compelling compliance, on pain of sanctions. 319 9 U.S.C. § 7. On the back end, if the agreement so provides, a party may seek a court order confirming any arbitral award, 320 Id. § 9. with the order “hav[ing] the same force and effect, in all respects, as, and be[ing] subject to all the provisions of law relating to, a judgment in an action,” such that “it may be enforced as if it had been rendered in an action in the court in which it is entered.” 321 Id. § 13. The FAA thus repeatedly endows the decisions of private parties and their chosen arbitrators with the full and virtually automatic coercive backing of the state 322 See Richard C. Reuben, Democracy and Dispute Resolution: The Problem of Arbitration, 67 Law & Contemp. Probs. 279, 296 (2004) (noting that arbitral awards “generally are final, binding, and enforceable by courts, and generally may not be reversed on substantive grounds”). —a delegation of state power.

What is more, in contrast to ordinary civil litigation, the FAA lacks any meaningful mechanisms for policing its delegations. In fact, the Act, as construed by the Supreme Court, affirmatively frustrates almost all judicial delegation-policing efforts, severely limiting the ability of courts to review a privately chosen arbitrator’s actions before coercively enforc­ing them. 323 See Horton, Delegation, supra note 7, at 489–93; Resnik, Diffusing Disputes, supra note 6, at 2810. Courts generally may not scrutinize an arbitrator’s subpoena for relevance or materiality before compelling compliance with it. 324 See, e.g., In re Sec. Life Ins. Co. of Am., 228 F.3d 865, 871 (8th Cir. 2000) (refusing to require district courts to “make an independent assessment of the materiality” of information sought in a subpoena before enforcing it); Festus & Helen Stacy Found., Inc. v. Merrill Lynch, Pierce Fenner, & Smith Inc., 432 F. Supp. 2d 1375, 1379–80 (N.D. Ga. 2006) (holding that determining the “relevance of the subpoenaed materials . . . should be left to the arbitration panel”). But see Oceanic Transp. Corp. of Monrovia v. Alcoa S.S. Co., 129 F. Supp. 160, 161 (S.D.N.Y. 1954) (holding that the district court had a duty to determine “whether or not the subpoenaed material would be material as evidence”). While the Act specifies various grounds for vacating, modifying, or correct­ing an arbitral award, 325 See 9 U.S.C. §§ 10–11. the Supreme Court has held that those grounds are exclusive and cannot be augmented, even by private agreement. 326 See Hall St. Assocs. v. Mattel, Inc., 552 U.S. 576, 584 (2008). But see Richard C. Reuben, Personal Autonomy and Vacatur After Hall Street, 113 Penn. St. L. Rev. 1103, 1110–16 (2009) [hereinafter Reuben, Personal Autonomy] (identifying various nonstatutory grounds for vacatur that State and lower federal courts had developed before Hall Street). Each of the enumerated grounds, moreover, is quite narrow. For example, most of the grounds for vacatur require egregious abuse during the proceedings, such as fraud or corruption, that actually infects the final award. 327 See 9 U.S.C. § 10(a)(1)–(3). Only one of the grounds concerns the merits of the award, permitting vacatur “where the arbitrators exceeded their powers, or so imperfectly executed them that a mutual, final, and definite award upon the subject matter submitted was not made.” 328 Id. § 10(a)(4). Yet, the Court has held that “[i]t is not enough” under this provision “to show that the [arbitrator] committed an error—or even a serious error”; 329 Oxford Health Plans LLC v. Sutter, 133 S. Ct. 2064, 2068 (2013) (alteration in original) (internal quotation marks omitted) (quoting Stolt-Nielsen S.A. v. AnimalFeeds Int’l Corp., 559 U.S. 662, 671 (2010)). a court must uphold an award so long as the arbitrator “even arguably constru[ed] or appl[ied] the [parties’] contract.” 330 Id. (internal quotation marks omitted) (quoting E. Associated Coal Corp. v. United Mine Workers, Dist. 17, 531 U.S. 57, 62 (2000)). The FAA, in short, requires federal courts to coercively enforce all but the most flagrantly abusive decisions of pri­vately appointed arbitrators. Given the statute’s broad delegations and dearth of delegation-policing mechanisms, it seems fair to conclude along with proponents of the delegation critique that the FAA has “aggrandized private parties.” 331 Horton, Delegation, supra note 7, at 456.

Scholars are right to worry about this state of affairs and to decry the Supreme Court’s seeming nonchalance toward the FAA’s broad delega­tions of state power. In addition to all of the concerns raised by propo­nents of the delegation critique, this Article has highlighted the risk that private parties will abuse delegations such as those in the FAA, violating important public values and potentially rendering the entire political community complicit in their wrongdoing. Perhaps we could tolerate that risk if, as in ordinary civil litigation, courts possessed meaningful mechanisms to repudiate abusive exercises of delegated power in arbitra­tion. But the Supreme Court has largely eviscerated what few delegation-policing mechanisms the FAA arguably contains.

2. Beyond Enforceability. — Proponents of the delegation critique have proposed to address this problem primarily by curbing arbitration—essentially, limiting private parties’ access to the FAA’s delegations of state power. They would do so in two ways. On the one hand, many scholars would have courts refuse to enforce arbitration clauses in broad catego­ries of disputes, particularly disputes between sophisticated parties and individual consumers or employees, mandating that such disputes be resolved exclusively through court-based adjudication. 332 See supra note 27 and accompanying text; see also, e.g., Linda S. Mullenix, Gaming the System: Protecting Consumers from Unconscionable Contractual Forum-Selection and Arbitration Clauses, 66 Hastings L.J. 719, 756–60 (2015); Resnik, Privatization, supra note 6, at 1812; Reuben, Constitutional Gravity, supra note 9, at 1017–34; Richard C. Reuben, FAA Law, Without the Activism: What If the Bellwether Cases Were Decided by a Truly Conservative Court?, 60 U. Kan. L. Rev. 883, 922–24 (2012). Scholars have made similar proposals for other forms of procedural private ordering, such as forum-selection clauses. See, e.g., Paul B. Carrington & Paul H. Haagen, Contract and Jurisdiction, 1996 Sup. Ct. Rev. 331, 332, 351–61, 401; Linda S. Mullenix, Another Choice of Forum, Another Choice of Law: Consensual Adjudicatory Procedure in Federal Court, 57 Fordham L. Rev. 291, 362–63 (1988). Because the arbi­tration clauses governing such disputes almost always appear in contracts of adhesion, scholars argue that courts should either refuse to enforce adhesive arbitration clauses outright or, to similar effect, rigorously assess the “voluntariness” of such clauses, defining “volun­tariness” in a way that precludes enforcement. 333 See supra notes 27, 332 and accompanying text. On the other hand, some scholars would have courts enforce only those arbitration clauses that provide for arbitration proceedings that closely mirror ordinary civil litigation. 334 See supra note 28 and accompanying text; see also, e.g., Resnik, Diffusing Disputes, supra note 6, at 2811, 2921; Resnik, Contract, supra note 6, at 600; Reuben, Constitutional Gravity, supra note 9, at 1046–101; Teresa J. Verges, Evolution of the Arbitration Forum as a Response to Mandatory Arbitration, 18 Nev. L.J. (forthcoming 2018) (manuscript at 5, 18–19, 22) (on file with the Columbia Law Review); cf. Sarah Rudolph Cole & E. Gary Spitko, Arbitration and the Batson Principle, 38 Ga. L. Rev. 1145, 1147–51 (2004) (considering the extension of civil Batson to arbitration); Paul R. Verkuil, Privatizing Due Process, 57 Admin. L. Rev. 963, 983–86 (2005) (arguing that arbitration should be shaped by due process values). Arbitration, on this view, would be much less disconcerting if it resembled the proverbial “day in court,” with ample discovery, an impartial adjudicator, and ade­quate remedies. To be sure, some of these same scholars simultaneously advocate enhanced judicial review of arbitral awards and proceedings—but only as part of a belt-and-suspenders approach rather than as a distinct strategy. 335 See, e.g., Resnik, Diffusing Disputes, supra note 6, at 2811 (advocating “oversight on both process and outcomes ex post” alongside various ex ante restrictions on enforceability); Stone, Rustic Justice, supra note 7, at 1025–28 (advocating due process protections, heightened review, and reinstatement of the revocability doctrine). And they ultimately put more weight on limiting the enforce­ment of arbitration clauses, as opposed to more closely scrutinizing the outputs of arbitration proceedings. 336 See, e.g., Horton, Delegation, supra note 7, at 493–96 (calling for greater judicial review of arbitration but emphasizing review of arbitration clauses as part of the enforceability and arbitrability analyses, rather than review of arbitral awards and proceedings as part of the vacatur and confirmation analyses). Scholars have recently begun to address these kinds of proposals to administrative agencies as well as courts. See, e.g., Daniel T. Deacon, Agencies and Arbitration, 117 Colum. L. Rev. 991, 993–94 (2017) (advocating greater administrative regulation of arbitration); David L. Noll, Regulating Arbitration, 105 Calif. L. Rev. 985, 1031–33 (2017) (same). But see Zachary D. Clopton, Class Actions and Executive Power, 92 N.Y.U. L. Rev. 878, 884, 887 (2017) (identifying drawbacks of an administrative approach).

This emphasis on the enforceability of arbitration clauses is under­standable as a practical matter. For at least as currently construed by the Supreme Court, the FAA gives courts somewhat more leeway to keep disputes out of arbitration in the first place than to reassert control over them once they’ve been shunted there—either by deeming an arbitration clause unenforceable under State and federal law doctrines preserved by the Act’s “savings clause” 337 9 U.S.C. § 2 (2012). The Court has, however, significantly cabined the savings clause, interpreting § 2 to preempt a number of State law doctrines, and to displace a number of federal law doctrines, that would keep more categories of disputes in court. See, e.g., Am. Express Co. v. Italian Colors Rest., 133 S. Ct. 2304, 2310–11 (2013) (declining to apply the federal effective-vindication doctrine); AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 346 (2011) (finding that California’s unconscionability doctrine “interferes with arbitration”). or as part of determining whether the parties actually agreed to arbitrate a particular dispute (the so-called arbitrability determination). 338 See Buckeye Check Cashing, Inc. v. Cardegna, 546 U.S. 440, 443–45 (2006) (discussing arbitrability). Unless, that is, the parties agreed to delegate that determination to the arbitrator. See Rent-A-Center, W., Inc. v. Jackson, 561 U.S. 63, 78 (2010) (allowing parties to delegate the arbitrability determination to arbitrators). For a critical overview of such “delegation clauses,” see generally David Horton, Arbitration About Arbitration, 70 Stan. L. Rev. (forthcoming 2018) (on file with the Columbia Law Review).

Scholars have also adduced deeply principled reasons for restricting access to arbitration’s delegations in broad categories of cases. First, sophisticated parties, such as corporations, frequently insert arbitration clauses into contracts of adhesion with ordinary individuals, a practice that, many believe, allows them to exploit the FAA’s delegations to effectively force individuals to relinquish their right to their “day in court” for a one-sided arbitration proceeding. 339 See, e.g., Radin, Fine Print, supra note 8, at 19–32; Horton, Delegation, supra note 7, at 486–89; Resnik, Diffusing Disputes, supra note 6, at 2808, 2839–40, 2863–74; see also Linda J. Demaine & Deborah R. Hensler, “Volunteering” to Arbitrate Through Predispute Arbitration Clauses: The Average Consumer’s Experience, 67 Law & Contemp. Probs. 55, 72–74 (2004); Knapp, supra note 308, at 788–89; David S. Schwartz, Mandatory Arbitration and Fairness, 84 Notre Dame L. Rev. 1247, 1249 (2009); Sternlight, Mandatory Arbitration, supra note 27, at 1648–53. Scholars have criticized other forms of procedural private ordering on similar grounds. See, e.g., Taylor & Cliffe, supra note 10, at 1105. Many have argued that the FAA was never intended to apply to such contracts but only to contracts between sophisticated commercial parties. 340 See, e.g., Circuit City Stores, Inc. v. Adams, 532 U.S. 105, 124 (2001) (Stevens, J., dissenting); Imre Szalai, Outsourcing Justice: The Rise of Modern Arbitration Laws in America 188–89 (2013); Christopher R. Leslie, The Arbitration Bootstrap, 94 Tex. L. Rev. 265, 300–20 (2015). Many proponents of the delegation critique sympathize with this argument, see, e.g., Resnik, Diffusing Disputes, supra note 6, at 2836–38, though their concerns about “privatization” and “outsourcing” do not depend on it. The exploitative nature of these agreements is exemplified for many scholars by arbitra­tion clauses that forbid class arbitration even in so-called negative-value cases, in which the costs of prosecuting the case on an individual basis (either in arbitration or a lawsuit) exceed any potential recovery. 341 For a cogent indictment of class-arbitration bans, see Italian Colors, 133 S. Ct. at 2313–20 (Kagan, J., dissenting); see also, e.g., Myriam Gilles & Gary Friedman, After Class: Aggregate Litigation in the Wake of AT&T Mobility v. Concepcion, 79 U. Chi. L. Rev. 623, 628–29 (2012).

Second, whereas most scholars tend to view arbitration’s delegations as differing from civil procedure’s in kind, this Article has essentially argued that the two differ in degree; arbitration and ordinary civil litigation may differ in the extent to which each delegates state power, but they do not differ in the fact that each delegates state power. Even accepting that characterization, however, one might think that the differ­ence in degree makes all the difference. Civil procedure delegates various procedural powers piecemeal while retaining as the final decisionmaker a public official who is constrained by both substantive law and procedural rules and who must give public reasons for her decisions; arbitration, by contrast, authorizes private parties to assign the state’s dispute-resolution function wholesale to a private decisionmaker.

Both sets of concerns have a lot of merit, but they have been fully aired in the literature, as have the resulting proposals to dramatically curtail arbitration. 342 See supra notes 6–9, 339 and accompanying text. Notwithstanding the extensive scholarly criticism, the Supreme Court has steadfastly adhered to its decades-old jurispru­dence upholding the enforceability of adhesive arbitration clauses against individual consumers and employees. 343 Cf. Markovits, supra note 153, at 487 (“Arbitration . . . is here to stay.”). What proponents of the delega­tion critique have missed is that civil procedure contains the seeds of another kind of response to arbitration’s delegation problem, one with its own, distinct logic. Even if courts continue to enforce arbitration clauses that many scholars consider problematic, we should empower courts to repudiate the abuses of delegated state power that occur during arbitration after the fact.

A delegation-policing approach holds promise notwithstanding the significant differences between arbitration and ordinary civil litigation. The power asymmetries between the parties do indeed tend to be more pronounced in arbitration than in civil litigation, and arbitration involves the wholesale delegation of the state’s dispute-resolution powers to privately chosen arbitrators, in contrast to the piecemeal delegations in civil litigation. In light of all this, many scholars worry that sophisticated parties can use arbitration to effectively suppress legal claims, making arbitral proceedings so one-sided that no victim of wrongdoing would bother to invoke them to seek redress. 344 See supra note 8 and accompanying text; see also, e.g., Resnik, Access, supra note 6, at 143–54 (reporting empirical findings that individual consumers rarely use arbitration). But adopting delegation-policing doctrines might still go at least some way toward addressing this concern. If courts reviewed arbitral awards and proceedings more rigorously before enforcing the awards—if arbitration’s delegations were better policed—the prospect of such review might have a feedback effect of rendering those processes less one-sided in many cases. It might then become more worthwhile for would-be plaintiffs to actually go through with an arbitration when their claims were subject to a valid arbitration clause. Greater policing ex post, in short, might help to make arbitration fairer ex ante.

To be clear, such ex post delegation-policing strategies may be insuf­ficient to resolve some of the biggest controversies surrounding arbitra­tion. It may be impossible, for example, to address the problems posed by class-arbitration bans without prohibiting them ex ante. Nonetheless, we can combat a significant amount of abuse by having courts police arbitration ex post, through analogues of civil procedure’s delegation-policing doctrines.

B. Policing Arbitration’s Delegations

This section begins to think about how we might recast civil proce­dure’s delegation-policing doctrines to fit arbitration’s delegations of state power. Given the magnitude of the task, the doctrinal proposals outlined here are necessarily impressionistic. This section aims only to give a gen­eral sense of what a delegation-policing approach to arbitration might look like rather than to develop a comprehensive program for reform. In doing so, however, this section attempts to illustrate how we can look to civil procedure to allay some of the concerns about arbitration.

1. Rescinding Delegations. — The FAA currently contains no cognates of the doctrines in civil procedure that rescind delegated state powers when they’re being abused. It seems, however, that such doctrines could play a salutary role in arbitration too. One possibility would be to amend the FAA to provide for interlocutory judicial review of certain procedural decisions made during an arbitration and for the termination of the arbitration if a court determined that any of those decisions were abu­sive. 345 It is important to distinguish this proposed review from interlocutory appellate review of district court decisions regarding arbitration. The FAA already provides for such appeals, though only from district court decisions refusing to enforce an arbitration clause. Compare 9 U.S.C. § 16(a)(1)(B) (2012) (permitting interlocutory appeals from district court refusals to compel arbitration), with id. § 16(b)(2) (forbidding most interlocutory appeals from district court orders compelling arbitration). The proposal might work as follows. Unless an arbitration clause was exploitative on its face, a court would compel arbitration upon the defendant’s request and stay any related judicial proceedings initiated by the plaintiff. But the plaintiff would be allowed to return to court during the course of the arbitration to challenge some of the arbitrator’s procedural decisions. If the court agreed with the plaintiff that the arbitrator had abused her delegated powers in making those decisions, it would discontinue the arbitration and lift the stay on the plaintiff’s lawsuit, thereby reasserting state control over delegated dispute-resolution powers.

This proposal would help to address some of the procedural deficiencies in arbitration identified by proponents of the delegation critique. As noted earlier, scholars have faulted arbitration for all too frequently denying plaintiffs the kinds of procedural protections they would enjoy in ordinary civil litigation, particularly a meaningful opportunity to obtain discovery and present evidence. 346 See supra notes 28, 334 and accompanying text. When an arbitrator denies those opportunities, and when the plaintiff can’t make her case without them, a court could deem the arbitrator’s decision an abuse of the state’s dispute-resolution powers and reassert state authority over the dispute. The bar for what constitutes “abuse” would, of course, have to be set quite high. It would not be sufficient for a plaintiff simply to disagree with an arbitrator’s determination that, say, a particular document wasn’t discoverable or the testimony of a particular witness was cumulative. But it seems reasonable to rescind delegated state powers from an arbitrator whose procedural decisions clearly thwart a fair resolution of the dispute. 347 The FAA does already authorize federal courts to vacate a final arbitral award when the arbitrator “refus[ed] to hear evidence pertinent and material to the controversy.” 9 U.S.C. § 10(a)(3). But courts have interpreted this provision narrowly. See, e.g., La. Dep’t of Nat. Res. ex rel. Coastal Prot. Restoration Auth. v. FEMA, No. 17-30140, 2018 WL 611215, at *2 (5th Cir. Jan. 29, 2018) (noting that vacatur under § 10(a)(3) is permitted only in “very unusual circumstances”). And in any case, a party shouldn’t have to await a final award before being able to challenge egregious abuse by an arbitrator.

This proposal prompts an obvious objection: Permitting interlocu­tory judicial review of arbitrators’ procedural decisions would invite a flood of piecemeal appeals, which would undermine the benefits of arbitration and burden the courts. This problem could be substantially ameliorated, however, by making interlocutory review discretionary with the court rather than as of right, 348 Cf. Fed. R. Civ. P. 23(f) (authorizing discretionary interlocutory appeals from class-certification decisions). and limiting such review to just a few key pro­cedural decisions, such as the decisions whether to permit discovery and the presentation of evidence. And courts could further discourage abuse of the interlocutory-review mechanism by imposing sanctions for review petitions that did not even plausibly allege an abuse of delegated power.

2. Withholding Enforcement. — The FAA already contains a few provi­sions that, properly construed, could provide a basis for implementing a strategy of withholding coercive enforcement from abusive exercises of delegated power in arbitration. For one, the FAA authorizes federal district courts to compel compliance with an arbitrator’s subpoena of a nonparty witness. 349 9 U.S.C. § 7. Although courts tend to rubber-stamp arbitral subpoenas, 350 See supra note 324 and accompanying text. one could imagine a regime in which judges scrutinized the subpoenas more closely and refused to compel compliance with those that abuse the state’s investigatory powers.

For another, the Supreme Court could overturn its precedents nar­rowly interpreting the provisions of the FAA governing the confirma­tion, vacatur, and modification of final arbitral awards and permit closer judicial review of such awards (though given its current composition, the Court is unlikely to change course anytime soon). 351 See 9 U.S.C. §§ 9–11 (specifying the procedures for confirmation, vacatur, correction, and modification of an arbitral award); Resnik, Diffusing Disputes, supra note 6, at 2882 (lamenting that the Supreme Court “has repeatedly stipulated the adequacy of arbitrations and rejected judicial monitoring of the outcomes”). Courts would not, of course, review arbitral awards de novo but would instead apply something more akin to the deferential “arbitrary and capricious” standard for judicial review of the decisions of administrative agencies. 352 The model would be the more deferential variant of arbitrary and capricious review. But cf. Richard A. Nagareda, Turning from Tort to Administration, 94 Mich. L. Rev. 899, 903 (1996) (advocating “hard look” review of class action settlements in mass-tort cases). Instituting even such limited review, however, would at the very least require overruling the Supreme Court’s narrow interpretation of 9 U.S.C. § 10(a)(4). See supra notes 327–330 and accompanying text. Some State and lower federal courts appear to hold out the possibility of arbitrary and capricious review for arbitral awards, but such “claims almost always fail.” Reuben, Personal Autonomy, supra note 326, at 1115. Another alternative would be to expand some courts’ practice of reviewing arbitral awards for a “manifest disregard of the law,” see id. at 1110–13, though understanding that standard to restrict arbitrators’ discretion more than it currently does, see id. at 1111 (noting that “few” courts actually set aside arbitral awards for manifest disregard). While many scholars have advocated greater judicial scrutiny of arbitral awards, this Article’s delegation framework grounds that proposal in a distinct logic rather than seeing it as just another way to curb arbitration.

Bolstering judicial review would go some way toward addressing a significant concern with arbitration: that most arbitrators are biased in favor of the repeat players that pay their fees and generate most of their business. 353 See, e.g., Reuben, Constitutional Gravity, supra note 9, at 1064–67 (describing the “repeat player phenomenon”); Schwartz, Claim-Suppressing Arbitration, supra note 306, at 248 (“[T]he system of arbitration as a structural matter creates financial incentives to decide questions of arbitrator jurisdiction (‘who decides’) and even merits issues favorably toward those who pay them.”); Jessica Silver-Greenberg & Michael Corkery, In Arbitration, a ‘Privatization of the Justice System,’ N.Y. Times (Nov. 1, 2015), (on file with the Columbia Law Review) (citing records showing that “41 arbitrators each handled 10 or more cases for one company between 2010 and 2014”). The Constitution requires an impartial arbitrator. 354 Commonwealth Coatings Corp. v. Cont’l Cas. Co., 393 U.S. 145, 150 (1968). But this requirement confronts a well-recognized dilemma: If a repeat-player defendant pays an arbitrator’s fees, the arbitrator may be incentivized to rule for the defendant, in the hope that it will continue to send disputes her way; yet a defendant can’t require an individual plaintiff to pay all of the arbitrator’s (considerable) fees, lest it effectively foreclose all relief. Notwithstanding this dilemma, scholars have argued that courts should refuse to enforce arbitration clauses that create even the potential for subtle forms of bias. 355 See, e.g., Reuben, Constitutional Gravity, supra note 9, at 1062–67 (calling for inquiries into “passive bias” and “structural bias”). Under a delegation-policing approach, by contrast, a court would review an arbitral award for arbitrariness or capriciousness. While that standard would still afford arbitrators significant discretion, it would filter out those proceedings in which the arbitrator was actually biased and the bias infected the award by prompting the arbitrator to rule based on partiality for the defendant rather than any legal or contractual ground. 356 Cf. Cuyler v. Sullivan, 446 U.S. 335, 349–50 (1980) (holding that, to prove a violation of the right to conflict-free counsel, a defendant must show “that his counsel actively represented conflicting interests”). The prospect of such review, moreover, could potentially deter many forms of bias. This proposal could be grounded in the provision of the FAA permitting vacatur based on “evident partiality or corruption in the arbitrators,” 357 9 U.S.C. § 10(a)(2). but understanding “partiality” and “corruption” more capaciously than courts currently do. 358 See, e.g., Positive Software Sols., Inc. v. New Century Mortg. Corp., 476 F.3d 278, 283 (5th Cir. 2007) (en banc) (“[I]n nondisclosure cases, an award may not be vacated because of a trivial or insubstantial prior relationship between the arbitrator and the parties to the proceeding.”). That broader conception seems justified once one recognizes that arbitrators are vested with significant state power and therefore have a responsibility to exer­cise that power consistent with important public values.

One practical hurdle this proposal would have to surmount is that there is no right to a reasoned opinion in arbitration, and arbitrators rarely provide one (though sometimes a transcript of the arbitrator’s decision is available). 359 See Reuben, Constitutional Gravity, supra note 9, at 1082–91 (canvassing arguments for and against reasoned opinions but recognizing that there is no constitu­tional right to a reasoned opinion). The lack of a reasoned opinion could frustrate judicial review in many cases, preventing a party from proving that the arbitrator or other party abused delegated state power. Short of requiring reasoned opinions in all arbitrations, courts could impose a “penalty default” 360 Cf. Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 91 (1989) (defining a “penalty default” as a default rule that is “purposefully set at what the parties would not want—in order to encourage the parties to reveal information to each other or to third parties (especially the courts)”). in the absence of such an opinion, shifting the burden to the party seeking confirmation or opposing vacatur to prove that the alleged abuse of delegated power did not in fact occur. This would likely have the effect of incentivizing parties to require their chosen arbitrators to prepare reasoned opinions, just as expanded judicial review in general would likely incentivize arbitrators to exercise their powers more dispassionately.

3. Punishing Abuse. — Finally, courts could punish abuse of delegated state power during arbitration by imposing sanctions during actions to confirm arbitral awards. The standard for imposing sanctions would resemble Rule 11’s “improper purpose” standard. 361 See supra notes 209–211 and accompanying text. According to such a standard, sanctions would be warranted when one party sought to confirm an arbitral award that issued from severely defective proceedings or sought confirmation after having forced the other party into arbitration for some illegitimate reason, such as to deprive her of an effective remedy for her injuries. 362 Cf. Carnival Cruise Lines, Inc. v. Shute, 499 U.S. 585, 595 (1991) (considering, in assessing a forum-selection clause for “fundamental fairness,” whether the defendant included the clause with the “bad-faith motive” of “discouraging [plaintiffs] from pursuing legitimate claims”). Courts could impose sanctions under either Rule 11 itself or their inherent powers. 363 See supra note 214 and accompanying text. States could also create a cause of action for “abuse of arbitral process,” lying against the party who compelled the arbitration or the arbitrator herself, depending on who was ultimately respon­sible for the abuse. 364 Creating a cause of action against arbitrators would require overturning decisions cloaking them in absolute judicial immunity. See, e.g., Austern v. Chi. Bd. Options Exch., Inc., 898 F.2d 882, 886 (2d Cir. 1990). See generally Maureen A. Weston, Reexamining Arbitral Immunity in an Age of Mandatory and Professional Arbitration, 88 Minn. L. Rev. 449 (2004) (arguing that arbitrators should enjoy only qualified, not absolute, immunity). By punishing abuse of arbitra­tion’s delega­tions, the state would not only repudiate the abuse but also incentivize private parties and arbitrators to design and conduct arbitra­tions in accordance with important public values.

* * *

The foregoing proposals are not without their problems. 365 So long as they were limited to purely domestic arbitrations, none of the proposals would put the United States in violation of its obligations under the New York Convention, which governs the enforcement of foreign arbitral awards. See Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York Convention), June 10, 1958, 21 U.S.T. 2157, 330 U.N.T.S. 38; see also 9 U.S.C. §§ 201–208 (2012) (prescribing procedures for the enforcement of foreign arbitral awards in federal court). This Article does not address the distinct issues presented by international arbitration. Those who are skeptical of arbitration will likely worry that the various dele­gation-policing strategies, with their reliance on ex post judicial review, would prove too burdensome for the parties they’re primarily designed to help—namely, individuals seeking redress against sophisticated parties. Yet while plaintiffs and plaintiffs’ lawyers may currently have inadequate incentives to challenge arbitral awards and procedures, the calculus would be different if the delegation-policing strategies outlined above were adopted. Indeed, with stronger delegation-policing doctrines, it might even be easier to challenge arbitral awards and procedures after the fact than to oppose sending a dispute to arbitration in the first place. That’s because the primary legal bases for resisting the enforcement of an arbitration clause—the State contract law doctrines of public policy and unconscionability and the federal effective-vindication doctrine 366 See supra note 337. —all turn on speculative inquiries about the likely conduct of the arbitra­tion proceedings in a given case or set of cases, whereas the delegation-policing doctrines respond to abuse that has already occurred.

Supporters of arbitration, on the other hand, will likely worry that the delegation-policing strategies would threaten to vitiate many of arbitra­tion’s purported benefits, such as finality, informality, and speed. 367 See, e.g., Lisa Bernstein, Understanding the Limits of Court-Connected ADR: A Critique of Federal Court-Annexed Arbitration Programs, 141 U. Pa. L. Rev. 2169, 2240–41, 2243 (1993); see also Hall St. Assocs. v. Mattel, Inc., 552 U.S. 576, 588 (2008) (explaining that limiting judicial review “maintain[s] arbitration’s essential virtue of resolving disputes straightaway,” whereas permitting “full-bore legal and evidentiary appeals” would render arbitration “merely a prelude to a more cumbersome and time-consuming judicial review process” (internal quotation marks omitted) (quoting Kyocera Corp. v. Prudential-Bache Trade Servs., Inc., 341 F.3d 987, 998 (9th Cir. 2003))). Some scholars have suggested that these benefits are exaggerated, as arbi­tration has increasingly come to resemble litigation. 368 See, e.g., Edward Brunet, Replacing Folklore Arbitration with a Contract Model of Arbitration, 74 Tul. L. Rev. 39, 52–61 (1999) (arguing that the rules of the major arbitration organizations have become highly judicialized); Thomas J. Stipanowich, Arbitration: The “New Litigation,” 2010 U. Ill. L. Rev. 1, 11–20 (2010) (arguing that arbitration increasingly employs litigation-like procedures, including discovery and hearings). But whatever the merits of that debate, the delegation-policing approach advocated here would likely undermine arbitration primarily in one-off disputes between unequal parties—precisely the kinds of disputes in which arbitration is most likely to involve abuse. For the weaker party in such disputes would have little reason not to invoke the delegation-policing doctrines. In disputes between sophisticated parties, by contrast, the parties would be more likely to have an ongoing relationship and thus would be more loath to invoke the delegation-policing doctrines, lest they undermine the efficiency advantages of arbitration for their future disputes. 369 Cf. Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. Pa. L. Rev. 1765, 1796 (1996) (identifying “norms that transactors follow in attempting to cooperatively resolve disputes in a manner that will not jeopardize future dealings”); Stone, Rustic Justice, supra note 7, at 1024–30 (arguing that arbitration is most appropriate for disputes between the members of “self-regulating normative communities”). It is also worth reemphasizing that, in any kind of dispute, the standard for showing an abuse of delegated power would be a strict one, such that the majority of arbitral decisions would in practice be final. In the end, though, we inevitably face a tradeoff between benefits to private parties and costs to public values whenever we delegate state power. A delegation-policing approach better accounts for both the benefits and the costs of arbitration’s delegations than does the current regime.


For solutions to some of the most vexing problems with arbitration, we should look to civil procedure. Scholars have rightly worried that arbitration delegates state power to private parties. What they’ve missed, however, is that a similar delegation problem has already been rehearsed in ordinary civil litigation. One of the most distinctive features of the American civil justice system is that it allows private parties to exercise significant coercive power without prior judicial approval; it is a relatively small step to conceptualize those private exercises of power as delega­tions of state power. Responding to the potential for its delegations to be abused, civil procedure contains numerous doctrines that, rather than restrict access to delegated power before any abuse has occurred, allow the state to distance itself after the fact from exercises of delegated power that offend especially important public values. Civil procedure has even articulated a normative rationale for these delegation-policing doctrines: the need for the state (and the political community it serves) to avoid complicity in private wrongdoing. So the fundamental problem with arbi­tration isn’t that it delegates state power per se, but that it lacks the kinds of delegation-policing doctrines found throughout civil procedure. Once we grasp this major flaw in the current arbitration regime, we can move beyond the dichotomy between permitting private parties to abuse arbitration’s delegations with virtual impunity, as current doctrine does, and putting strict ex ante limits on those delegations, as most critics of arbitration advocate. Even if arbitration clauses continue to be enforced more often than arbitration’s critics think legitimate, we can still address many of the concerns about arbitration by expanding judicial review of arbitral awards and proceedings to enable courts to police arbitration’s delegations, just as they already police civil procedure’s.

Why have scholars overlooked this deep continuity between arbitra­tion and ordinary civil litigation? This Article concludes by venturing a possible diagnosis and a prescription.

The diagnosis is the near monopoly procedural due process enjoys on normative questions in civil procedure. An analytical framework that is self-consciously preoccupied with the procedures of state institutions is bound to neglect crucial issues when we’re assessing the exercise of coer­cive power by private parties. Notwithstanding this blind spot, scholars continue to analyze procedural questions primarily in due process terms. 370 See, e.g., Resnik, Contract, supra note 6, at 663–64 (“Contract Procedure requires some oversight, through the lens of Due Process Procedure, because court-based decisionmaking entails opportunities to be heard.”); see also supra notes 28, 334 and accompanying text (advocating the extension of due process norms to arbitration). That approach will prove at best incomplete, eliding important issues raised by the privatization of the state’s dispute-resolution powers. For the same reasons, procedural due process also glosses over core fea­tures of ordinary civil litigation, which casts the state in a supporting role as often as a leading one.

The prescription, as this Article has sought to demonstrate, is a greater focus in civil procedure scholarship on liberal political theory and its account of political power. Scholars commonly turn to liberal political theory to understand other areas of the law, particularly consti­tutional law 371 See, e.g., Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787, 1796–1801 (2005) (discussing “moral and political legitimacy” as an aspect of “constitutional legitimacy”); Richard H. Pildes, Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism, 27 J. Legal Stud. 725, 726 (1998) (exploring the connections between constitutional law and theories of rights). and criminal law. 372 See, e.g., George P. Fletcher, The Nature and Function of Criminal Theory, 88 Calif. L. Rev. 687, 700 (2000) (“This is fairly good proof that criminal theory seems to function today as a potpourri of political theory.”); V.F. Nourse, Reconceptualizing Criminal Law Defenses, 151 U. Pa. L. Rev. 1691, 1703–26 (2003) (analyzing criminal law defenses in terms of political theory). And it has even made some significant inroads into civil procedure, as scholars have invoked liberal-democratic values to justify various aspects of civil litigation. 373 For a recent example, see generally Alexandra Lahav, In Praise of Litigation (2017). Under the early influence of Lea Brilmayer, personal-jurisdiction scholarship has also incorporated many insights from political theory. See, e.g., Lea Brilmayer, Liberalism, Community, and State Borders, 41 Duke L.J. 1, 1 (1991) (analyzing personal jurisdiction in terms of political theory); George A. Rutherglen, Personal Jurisdiction and Political Authority (Va. Pub. Law & Legal Theory Research Paper No. 2017-13, 2017) (same), (on file with the Columbia Law Review). And Martin Redish has criticized the contemporary class action from a similar perspective. See generally Martin H. Redish, Wholesale Justice: Constitutional Democracy and the Problem of the Class Action Lawsuit (2009). But liberal political theory’s account of political power has been virtually absent from debates about the privatization of civil justice. That is unfortunate. Civil proce­dure is one of the primary ways political power gets exercised in our society. As such, it raises all the same questions that attend the exercise of political power in any other institutional setting—the boundary between public and private foremost among them. At the same time, civil proce­dure inflects these age-old questions with its unique division of labor between private parties and state officials. Liberal political theory can thus help us to more fruitfully conceptualize some of the problems with recent procedural trends, as well as to more readily apprehend the solutions our civil justice system already has to offer.